The Open Source Movement: Implications for Education

Comprehensive Essays

By Alec Couros Spring/Summer 2004

2

Foreword:

re: Content To have written ‘comprehensively’ on any subject is difficult with or without limits. Similarly, to write comprehensive papers regarding such complex and diverse topics as I have before me, and to limit myself to 20 pages per topic, is certainly a formidable task. Obviously, I have had to make choices on what content I should include, and what I should not. And these choices were never easy. However, I am hoping that these essays will serve as a launching point into describing what I know now, and what I hope to learn in the future. The state of the open source movement and the activities that arise from it have shifted and continue to shift before me as I write these essays. While these ideas seem to be constantly reshaping, they are no less interesting to me than they were at the beginning of this journey. The open source movement has become more than just a better way of writing software. The open source movement can be seen as a culture, an ideology and a better way for humans to work together on shared pursuits. re: Format This entire paper was written using open source and free software. Open Office Write (part of the Open Office Suite) was used as an alternative word processor to Microsoft Word. LaTex was used as the typesetting tool for the purpose of transforming the raw, unformatted document into a consistent format and in converting the finished text into PDF form. Additionally, open source tools are being used to serve this document to an online audience at http://www.educationaltechnology.ca/couros. The operating system of the host server is Linux-based, the web server is Apache and the content management system is driven by Moveable Type. Again, all of these tools are either open source or available for free. However, I don’t believe that these facts alone should be utterly remarkable. In the past several years, open source tools have become more widely accepted by business groups, educational organizations and consumers, and open source thought is slowly emerging into mainstream culture. Yet, what I do believe to be remarkable is the process in which such tools came to be. Through often informal and spontaneous cooperation, viable alternatives to the products of proprietary giants (e.g., Microsoft, Novell, Oracle) have brought forth a revolution of choice, quality and customizability. And, what is perhaps more important here is the understanding and acknowledgment of the collaborative and creative processes, and spirit, that brought about these changes. It is my hope that in better understanding the complexities of the open source movement, insight can be drawn for the purpose of developing collaborative, innovative communities within the educational context.

3

Question #1:

1) What is the open source movement?

a) Discuss the attributes and implications for educational practice and preparation.

b) Describe the sources and nature of the data that might be collected regarding this movement that might be found on publicly accessible listservs, discussion

groups and other Internet-based communities.

4

Starting at the “Source”

Before the term “open source” can be understood as the foundational idea

behind a socio-political movement, it should be first described at the fundamental

technical level. It is my assumption that before one can begin to understand

open source in the larger context, it is important to have a basic comprehension

of the term “source” and to understand other essential terms such as hardware,

software and programming languages. While one may argue that such

information is best left for an appendix, I believe that a basic understanding of the

technical aspects of open source will allow readers to better understand the

context and history of the open source movement.

In acknowledging the technical characteristics of a movement which

originates from early computing, there is an important distinction to be made

between the terms ‘hardware’ and ‘software’. Hardware is typically described as

including the physical components of a computer (e.g., keyboards, monitors,

hard drives, etc.). These items are usually fixed and less modifiable (by end

users) than software. Software, in contrast, includes programs, codes and

routines that control the overall operation of the hardware components. Software

programs, which include subsets of codes and routines, are most easily written

through the assistance of programming languages. A programming language is

an artificial language that assists in enabling programmers (human software

developers) to write instructions or codes that, once compiled, will be executed

by the hardware. Examples of today’s popular programming languages include

C++, Java and PERL.

5

The relationship between hardware and software is also important.

Programming languages allow a bridge between the human developer and the

control of the hardware, but written code from any programming language must

be compiled or translated to binary machine code (0’s and 1’s) for the hardware

to ‘understand’ and execute the commands. In other words, code written by

human programmers must be converted to binary code, and once this conversion

is complete, the code is only readable by computers.

Source or source code includes the actual text written by programmers in

any number of programming languages. To other programmers, this code is

readable and understandable. When code is in the source form, programmers

are best able to understand how code works, how code may be improved or as

often the case, how errors in code can be addressed. However, once the source

code is compiled to machine language, it is very difficult or impossible to work

with the code to make alterations or improvements to software (see Figure 1

below).

6

Figure 1: Steps in the conversion of human thought to machine code. Unless developers make source code available, the code is almost always impossible to retrieve, modify or improve.

Today, most software is either developed as proprietary (closed source) or

as open source. For instance, Microsoft Windows XP (and all other previous

versions of Windows) is completely closed source. Only developers at Microsoft

(or Microsoft designated third parties) are able to make any modifications to this

operating system (software). Alternately, many distributions of the increasingly

popular Linux operating system are available free of charge by download from

various Internet sources. Due to the potentially drastic reduction in cost of

systems running open source operating systems (e.g., Linux) versus proprietary

operating systems (e.g., Microsoft XP), the open source economic model has

begun to find favour with educators in both small and large-scale initiatives (see

7

examples of School Division initiatives, notably the State of Maine and Kamloops

School Division #73 at http://www.canopener.ca)

Origins of the Free/Open Source Movement

Although the term “Open Source Software” has only recently claimed

widespread use, this phrasing originates in the American “hacker” culture of the

1960’s and 70’s. More specifically, the practices behind the term are attributed to

key individuals working in the computer science laboratories of Stanford,

Berkeley, Carnegie Mellon and MIT. Such individuals include Eric Allman, the

inventor of “sendmail”, the open source program that is in part responsible for

email to move from sender to receiver (you may not have heard of Allman, but if

you’ve used email, you’ve used his invention). Ken Thompson, who moved from

Berkeley to work with Bell Labs in 1969, is also another key individual and the

inventor of Unix.1 Dennis Ritchie, who worked with Ken Thompson on the Unix

project, is attributed to developing C, one of the most widely used programming

languages. While these individuals may not be household names, their work in

computer science paved the way for the modern day connectivity we experience

in our personal, academic and business class computing.

These individuals represent a few major developments of the era which

hold continuing effects on modern computing. Yet, such advancements may not

have been possible had it not been for the collaborative nature of the developing

community. This pioneering community of programmers was small and close-

1 Unix is still the most common multi-user operating system in the world and the basis for Linux and other popular open source (e.g., BSD-based Unix varieties) and proprietary (e.g., SCO Unix) operating systems.

8

knit. Throughout the software development process, code (source) passed freely

between members on various projects. If an improvement was made to code,

there was an expectation that this information was to be passed along to other

members of the entire programming community. To withhold code was

considered gauche, as it was to everyone’s benefit in the collaborative culture

that the code was improved (Stallman 2000). However, the nature of this

collaborative protocol was to take a turn by the end of the 1970’s as many

developers were enticed to join commercial firms producing proprietary software.

Certainly the most profound and outspoken member of the Open Source

Movement is Richard Stallman who began his computer science career as a

Graduate student at MIT in 1971. While Stallman’s career began in an

environment of collaboration, sharing and collegiality (as described above), his

surroundings began to change in the early 1980’s as many of his former

colleagues began to work for commercial companies, which sold primarily

proprietary systems. In an interview with David Bennahum, Richard Stallman

spoke about the origins of this practice as he recalled the actions of student

programmer Brian Reed from Carnegie Mellon University in 1980. Reed, a

computer science student who wrote a text-formatting program named Scribe,

“surprised everyone by selling it to a company, instead of sharing it with the

community. The company was very proprietary about it, and very obnoxiously put

time bombs into it” (Bennahum, 1996, online). “The problem was that nobody

censured or punished this student for what he did. He got away with it. The result

was other people got tempted to follow his example” (King, 1999, online).

9

As ranks of programmers moved toward commercial pursuits, Stallman

struggled to maintain a community of collaboration. Stallman disclosed, “I was

faced with a choice. One: join the proprietary software world, sign the

nondisclosure agreements and promise not to help my fellow hackers. Two:

leave the computer field altogether. Or three: look for a way that a programmer

could do something for the good. I asked myself, was there a program or

programs I could write, so as to make a community possible again?” (King, 1999,

online). It seems that Stallman chose to follow his third choice as he left MIT in

1985.

Shortly after leaving MIT, Stallman pursued his lost community and

formed the Free Software Foundation (FSF). The foundations’ website

(http://www.gnu.org) hosts community-developed, freely available software. In

1985, Stallman authored the GNU manifesto2, which called for other

programmers/hackers to join a new project, which could potentially champion the

benefits of sharing source code. GNU, perhaps Stallman’s most famous

initiative3, is the most revered/used software package available from the

foundation. GNU, a recursive acronym, stands for “GNU’s not Unix”. GNU is not

one program but a collection of many complementary programs (e.g., compilers,

editors, text formatters, mail applications, etc.) created through the GNU project.

GNU is the combined work of many programmers and developers, and was

developed to be upwardly compatible with Unix. However, GNU was missing one

key component to make it a complete operating system; the kernel. In other

2 http://www.gnu.org/gnu/manifesto.html 3 http://www.gnu.org/

10

words, GNU had all the separate components of a complete operating system,

but was missing the central component that would mesh these parts together.

Enter Linus Torvalds and Linux.

The Rise of Linux and Other Open Source Giants

In 1991, Linus Torvalds, a 21-year-old computer science student at the

University of Helsinki posted the following message to the Internet newsgroup

comp.os.minix.

> From: torvalds@klaava.Helsinki.FI (Linus Benedict Torvalds) > Newsgroups: comp.os.minix > Subject: What would you like to see most in minix? > Summary: small poll for my new operating system > Message-ID: <1991Aug25.205708.9541@klaava.Helsinki.FI> > Date: 25 Aug 91 20:57:08 GMT > Organization: University of Helsinki > > > Hello everybody out there using minix - > > I'm doing a (free) operating system (just a hobby, won't be big and > professional like gnu) for 386(486) AT clones. This has been brewing > since april, and is starting to get ready. I'd like any feedback on > things people like/dislike in minix, as my OS resembles it somewhat > (same physical layout of the file-system (due to practical reasons) > among other things). Although seemingly an insignificant request, Torvald’s posting, and others like it

that followed, spurred the collaborative creation of what was to become the Linux

kernel, the central component of the Linux/GNU operating system.4 Although

Linux, as software, has been revered as a viable (and free) alternative to

Windows, what I believe to be more significant lies in the collaborative manner in

which Linux was developed. Raymond (1999) writes, “The most important feature

4 Note: Technically, Linux is not a complete operating system in itself, but refers to the kernel. The kernel is an essential part of the operating system, however, and is the core that provides basic services for all other parts of the operating system. Without the Linux Kernel, GNU was an incomplete operating system. For an excellent technical description of how Linux and GNU coexist, see Fink, 2003, Chapter Two.

11

of Linux, however, was not technical but sociological. Until the Linux

development, everyone believed that any software as complex as an operating

system had to be developed in a carefully coordinated way by a relatively small,

tight-knit group of people. This model is typical of both commercial software and

the great freeware cathedrals built by the Free Software Movement in the 1980’s”

(p. 16). Additionally, even those individuals closest to the “hacker” culture could

not themselves predict the great influence that Linux, and as an extension the

open source movement, would have on the computing world. To the latter point,

Raymond (1999) continues, “Linux is subversive. Who would have thought that

even five years ago (1991) that a world-class operating system could coalesce

as if by magic out of part-time hacking by several thousand developers scattered

all over the planet, connected only by tenuous strands of the Internet?” (p. 17).

Only two years after the release of Linux, groups such as Red Hat,

Debian and SuSE emerged to modify and improve, to give away, and in some

cases, sell their own distributions of Linux. Features continue to be added to

Linux, including SAMBA5, which allows Linux to transparently share files and

printers over even Microsoft-based networks. On the desktop (consumer level), it

is reported that Linux has reached almost a two-percent market penetration. This

may seem an insignificant proportion until it is compared to veteran Apple

Computer’s penetration which has been estimated to be anywhere between two

and eleven percent (Fink, 2003, p. 31). Today, Linux runs on at least 15 different

computer platforms including IBM mainframes, Macintosh, PC’s, Sun Systems

5 http://www.samba.org/

12

and Palm Pilots. Linux boasts over 10 million users and the number continues to

grow (Wheeler, 2002).

Linux is certainly not the only great success story of the open source

movement. In January 1998, Netscape, in a shocking move6, announced that it

would release its popular web browser suite, Netscape Communicator, as an

open source product. While there were various opinions on what seemed an

insane move by a corporate giant, the action introduced the idea of free software

as a legitimate idea to the business community. Netscape felt that it could tap

into a community of developers to improve its own product. By 2002, Mozilla7

was released as the first open source, production version of Netscape’s code.

Also, Apache8 web server, another open source application, is currently

the most widely used web server on the World Wide Web. Other applications

like Perl9 and PHP10 have become increasingly popular and essential tools for

many web programmers. IBM, Dell and Compaq now all provide servers that run

Linux. Apple Computers has based its newest Operating System on BSD, an

open source Unix-like operating system developed at the University of Berkeley.

There are literally thousands of applications that are available as open source,

and many are designed for business, personal computing or education. A

comprehensive list of open source projects can be found at SourceForge.net.11

6 This is sometimes referred to as the “shot heard ‘round the world” (Raymond,1999, p.203) 7 http://www.mozilla.org 8 http://www.apache.org 9 http://www.perl.org/ 10 http://www.php.net/ 11 http://sourceforge.net/

13

Open Source Licensing Structures

The “magic” which helped to establish Linux and, by now, countless

other open source and “free” software projects would have likely not occurred

without Stallman’s unique twist on copyright and software licensing. In the motion

picture documentary, Revolution OS, Torvalds described his relationship with

Stallman stating, “Think of Richard Stallman as the great philosopher and me as

the engineer.” To give more breadth to this statement, it’s important to

understand the General Public License (GPL), the idea of copyleft and Stallman’s

understanding of ‘free’ software.

The General Public License is the legal software agreement, which

enabled various developers to work on the GNU project. The detailed GPL and

its legalese can be found at the Open Source Initiative (OSI)12 website

(http://www.opensource.org/licenses/gpl-license.php). However, to summarize,

the GPL is based on Stallman’s following principles. When forming the Free

Software Foundation, Stallman stated that for any software to be truly “free”,

every user must have the right to:

1) run the program for any purpose. 2) modify the program to suit their needs (i.e., in most cases, this

means that potential developers must have access to the source code).

3) redistribute copies, either gratis or free. 4) distribute modified versions of the program, so that the

community can benefit from your improvements. (Free Software Foundation, Online)

From the previous set of principles, it may seem contradictory to say that

one can charge a fee for “free” software. To clarify, Stallman defines the meaning 12 Open Source Initiative (OSI) is a non-profit corporation dedicated to managing and promoting the ideals of open source software definition.

14

of the term free software by explaining that it refers to software that is “free as in

speech, not as in beer” (Dibona, Ockman, Stone, 1999, p.3). In other words,

Stallman is making a distinction between liberty and gratis. Software developers

of programs falling under the GPL have the liberty to make changes, share code,

use and redistribute, however, they are not bound to “give away” any derived

works. However, the notion of copyleft becomes very important here. Copyleft,

an essential part of the GPL is the mechanism that keeps software “free”.

Copyleft uses copyright law, but flips it over to serve the opposite of its usual purpose: instead of a means of privatizing software, it becomes a means of keeping software free. The central idea of copyleft is that we give everyone permission to run the program, copy the program, modify the program, and distribute modified versions – but not permission to add restrictions of their own. Thus, crucial freedoms that define “free software” are guaranteed to everyone who has a copy; they become inalienable rights. (Stallman, 2000, p. 53)

The GPL is not the only licensing agreement that applies to open source

software. In fact, GPL is only one of many open source-type licenses. However,

GPL is probably the most popularly referred to licensing structure due to

Torvalds’ decision to apply the GPL to Linux. Other popular open source

licenses include Apache Software license (the licensing structure of the world’s

most distributed web server software), Artistic License (applies to the Perl

programming language), BSD (Berkeley Systems Distribution)13, IBM public

license (IBM’s commercial copyleft license) and the Mozilla Public License

(applies to the open source version of the colossal Netscape Navigator) (Fink

2003). While most of these licensing structures are similar, each handles

intellectual property and the rights of the developers slightly differently. 13 Derived from a version of Unix developed at the University of Berkeley, California.

15

Developers have the ability to choose which licensing structure (in some cases,

more than one can be chosen) is most suitable to their work and to future

modifications done by the open source community or by commercial enterprise.

While there are certainly many variables in regards to the licensing of open

source, both initial developers of open source projects and those who adopt open

source software (e.g., schools) must be aware of the specific legalities in respect

to open source and free software. (see McGowan, 2001)

The Open Movement Today

When I first began thinking about the open source movement, I knew

only of the movement as it pertained to software. However, as I did more

research, I realized that software, while perhaps the impetus, was merely the

context. With closer scrutiny, what I found to be more important is that the open

source movement nurtures a philosophy that values certain freedoms pertaining

to knowledge dissemination, and leverages unique collaborative processes for

the creation of new knowledge. This basic philosophy has taken new shape

today as the open source movement has emerged, most notably, in the forms of

open content and open publishing.

Open content is an evolving term, but in a strict sense refers to

traditional and non-traditional content (e.g., books, articles, images, websites,

music, etc.) that is “freely available for modification, use, and redistribution under

a license similar to those used by the open source / free software community”

(Wiley, 1998, online). There has been the development of large open content

16

initiatives over the past several years. Notably, the Massachusetts Institute of

Technology (MIT) is currently developing the Open Courseware Initiative (OCW)

at http://ocw.mit.edu. The OCW is an open publication of MIT course materials

which currently features over 500 separate courses. The courses are licensed

under a Creative Commons14 license.

The Creative Commons (CC) is an organization that was developed to

help authors release content in a manner that protects the rights of the creator,

but also encourages accessibility toward certain public uses of material. The

developments of the Creative Commons are inspired by the open source

software movement and the organization has begun to redefine how authored

content is to be released. An excerpt from the CC philosophy reads:

We use private rights to create public goods: creative works set free for certain uses. Like the free software and open-source movements, our ends are cooperative and community-minded, but our means are voluntary and libertarian. We work to offer creators a best-of-both-worlds way to protect their works while encouraging certain uses of them — to declare "some rights reserved.” (Creative Commons, 2002, online)

Hundreds of works including books, essays, photographs, songs and

short videos have been released under Creative Commons’ licenses in the past

several years. The CC has become an attractive alternative apart from traditional

approaches of releasing content. And as the GPL licensing structure was

significant in fostering the collaborative development of Linux, certain CC

licenses can act to encourage and enable others to collaborate on joint projects,

while protecting attribution rights of the initial author(s). (To better understand the

14 http://creativecommons.org

17

CC licensing process, it’s helpful to view their “Choose License” page at:

http://creativecommons.org/license/)

Open publishing is also a relatively new phenomenon which has grown

under the influence of the open source movement, and has been encouraged by

the interactive structure of the Internet. Arnison (2001) writes, “open publishing

means that the process of creating news is transparent to the readers. They can

contribute to a story and see it instantly appear in the pool of stories publicly

available,,,, Readers can see editorial decisions being made by others. They can

see how to get involved and make editorial decisions” (Online). Transparency is

a key concept here, as the creation of news becomes open to the eyes of the

interested and involved, much like source code is accessible through open

source software projects. In the open publishing model, the relationship between

news producer and news consumer has merged.

In some instances, open publishing networks (often referred to as

Indymedias) seem to have risen as an intentional departure from the news

construction process at traditional media outlets. Many view the rise of open

publishing as a reaction against special interest, government influenced,

advertising-funded media empires (Hyde, 2000). In contrast, “Indymedias are

restructuring the traditional news hierarchy of publishers, advertisers, sources,

journalists and readers. In the world of Indymedia news, the relationship between

the sources, journalists, and readers is all that matters. In the Indymedia

community, publishers advertisers, and corporate interests are left out of the

picture” (Hyde, 2000).

18

There are dozens of examples of large open publishing ventures. Some

of the more popular examples include:

- The Independent Media Centre (IMC) - http://www.indymedia.org/. IMC is a collective of independent media organizations started in late 1999. There has been criticism of IMC, especially in regards to the anti-corporate and leftist slant to many of its stories.15

- TearItAllDown.com - http://www.tearitalldown.com/. An open publishing

community for varied types of activists. - Kuro5hin - http://www.kuro5hin.org/. A community focusing on

“technology and culture”, where daily users are able to vote for the days top stories, and comment on each.

- Slash Dot - http://slashdot.org/. Slash Dot’s motto is “news for nerds,

stuff that matters”. Daily news, especially related to technology, is posted daily and commentaries are written by users.

Another equally important development in the open publishing spectrum

includes the recent emergence of web logs (blogs). Blogs are usually personal

websites that provide updated journal entries, commentaries, headlines, news

articles or hyperlinks of interest to the user or to a somewhat specific audience.

While blogs have been around since early 1999, their rapid growth was fostered

more recently in part by the development of easy-to-use interfaces created by

Pitas 16 (the first web log specific software) and by Pyra Labs17 (known as

Blogger). Since then, dozens of content management systems18 have emerged

(e.g., Moveable Type, TypePad, Live Journal, Radio Userland, etc.) and there

are likely several million active bloggers.19 Blogging and content management

15 http://en.wikipedia.org/wiki/Independent_Media_Center 16 http://www.pitas.com 17 http://www.blogger.com 18 A content management system (CMS) is software that enables simple addition/editing/manipulation of HTML (website) content. 19 Blogspot alone has over 1.5 million active bloggers as of May 2004 (http://www.dijest.com/bc/).

19

systems are likely to have significant effects on the personalization and

practicality of publishing to and with others on the World Wide Web.

Rationale for Open Source Adoption

In the last several years, open source and free software development

has gained much popularity with educators (see Linux Case Studies20). Reasons

for open source advocacy vary, and dozens of informal articles by educators and

“techies” alike mark the World Wide Web eliciting varying rationales for open

source adoption (see Canopener21). Many argue economic reasons such as a

lower total cost of ownership (TCO) in schools22 and other, mostly economic

rationales (Lerner & Tirole, 2000). Others focus their arguments on social

democratic ideals of intellectual property (Newmarch, 2001, May, 2000, Moglen,

1999, Sodenberg, 2002). To use the colloquial term, it seems that “everyone and

their dog” has an opinion on open source software and Linux. However, the

difficulty I have encountered thus far is in finding peer-reviewed publications on

the topic, particularly related to how schools and educators are using open

source in the educational setting. Inquiries in this area could lead to identifying

what factors compel educators to use and or develop open source applications,

and simply, what is being done with open source and free software in schools?

A few theorists have compared the Internet economy, and particularly

open source communities, to that of the “Gift Economy” which was practiced

traditionally by North American First Nation tribes and ancient Chinese societies

20 http://casestudy.seul.org 21 http://www.canopener.ca 22A list of articles regarding TCO is available at http://luminance.sourceforge.net/resources_archive.php

20

(Cheal, 1988, Mauss, 1990). Partial gift economies also seem to exist, in

contemporary forms, within academic cultures. Barbrook (1998) writes,

Within small tribal societies, the circulation of gifts established close personal bonds between people. In contrast, the academic gift economy is used by intellectuals who are spread across the world. Despite the anonymity of the modern version of the gift economy, academics acquire intellectual respect from each other through citations in articles and other forms of public acknowledgement. Scientists therefore can only obtain personal recognition for their individual efforts by openly collaborating with each other through the academic gift economy. Although research is being increasingly commercialised, the giving away of findings remains the most efficient method of solving common problems within a particular scientific discipline. (Online)

Here, Barbrook both acknowledges and advocates for the collaborative

benefits of a gift culture.

Raymond (1998) also describes “hacker culture” as a true semblance of

gift economy. In Homesteading the Noosphere, he argues that our contemporary

society holds ties to a classical exchange society, which is strongly interlinked to

a command hierarchy. Added on to this, within our own contemporary society,

humans continue to hold an innate drive for social status. Using these

presumptions, Raymond (1998) proposes that “gift cultures are adaptations not

to scarcity but to abundance” (online). He continues, “Abundance makes

command relationships difficult to sustain and exchange relationships an almost

pointless game. In gift cultures, social status is determined not by what you

control but by what you give away.” Although Raymond may present an all too

simplistic analysis of the open source culture through this lens, there may be

aspects of the gift culture analogy that are worth examining. In particular, the gift

21

economy theories may in part help to explain motivations underlying the

increased popularity of open source software.

While there is much written on open source communities, especially

from an economic and technical (development) viewpoint, there is little written on

open source communities that involve educators. For this, I feel it necessary to

first identify educators that have or are beginning to adopt open source/free

software into their institutions, to examine their personal motivations for adopting

open source/free software into their personal life or practice and to begin to

examine how these educators are informed of issues with open source/free

software (technical, political, social) and learn to become competent users of

these tools. It is the assumption that these educators are members, at varying

levels, of open source communities, and therefore, it is important to the research

that their membership and participation in these (often virtual) communities is

identified and analyzed through an appropriate lens.

Informants and Sources of Data

As the context of this research is based on a relatively technical area,

traditional and non-traditional forms of data collection will be explored. The

following is a rough summary (pre-proposal) of how data collection methods may

be devised.

1) Selection of informants for face-to-face interviews, email conversations,

video conferencing, and other electronic communications. The desired

participants for this study are distributed over relatively large distances. While

some face-to-face interviews are possible, these will be limited because of

22

distance and time. Instead, electronic communication software is likely to be

used, that could include basic email, two-way audio or video conferencing,

listservs or web forums, blogging and possibly the use of wikis. Due to the

perceived technical abilities of the target group, I feel that communication over

distances using electronic tools will result in thoughtful, and rich data.

2) Tapping into existing open source-related listservs, forums, newsgroup,

blogs and other electronic communications There are many local, national

and international organizations (both formal and informal) that use electronic

communications that are either archived, or can be tracked for the duration of the

study. For instance, LOSURS (Linux Open Source Users of Regina) carries

active listserv conversations which often feature relevant information and/or

advocacy activities regarding open source thought and implementation in the

local context. In studying such electronic communications, two things may be

accomplished. First, communicators can be identified (by relevance and

experience) from the listservs and asked to participate in one-to-one interviews

or participate in face-to-face or electronic focus-groups. Second, the threads

themselves can be mined for rich data. Of course, either of these activities would

have to be approved by ethics, and in the case of the latter, by the

administrators/users of the listserv.

23

3) Publicly accessible open-source oriented organizations websites can be

mined for relevant content and activities. There are many open source related

websites that can be mined for rich data. Some of these websites include:

- CanOpener (http://www.canopener.ca) – Canopener is a national advocacy group which intends to educate Canadians on all aspects of free software, the open source movement and the use of GNU/Linux. - Free Software Foundation (http://www.fsf.org) - The website of Richard Stallman’s organization which presents a philosophy for free software, and fosters free software projects. - SchoolForge (http://www.schoolforge.net) – Schoolforge is a support website for schools hoping to implement open source initiatives into schools and districts. Additionally, many open source projects are developed through Schoolforge, and it’s sister organization, Sourceforge. - SEUL/EDU (http://www.seul.org/edu) – Simple End User Linux in Education is an organization dedicated to supporting the use of Linux in the educational context. - Open Source Initiative (http://www.opensource.org) – A non-profit corporation founded by Eric Raymond to promote Open Source initiatives through the development of standardized open source definitions. These are just a few of the important organizations that will act as contact

platforms, and potentially relevant sources of data. Together, these organizations

provide a wider perspective of the open source movement, in it’s many shapes

and forms, in Canada.

There are many potential sources of information for a study such as this.

However, due to the nature of the content, much of the research may rely on

information sources which are emergent and possibly non-refereed. While this

may cause “issues” to arise, as a researcher, I must consider the most relevant

and rich material, in whatever form. Then, through balancing and coalescing

these information sources with peer-reviewed literature and thoughtful

24

methodologies, I will be able to better understand the implications for the open

source movement in the educational context.

25

Question #2:

2) What is grounded theory and how will it support the work that you are

considering in your doctoral research?

a) What are two other qualitative approaches you considered, and rejected, and

why were they inferior to grounded theory in this instance?

26

Introduction to Grounded Theory

Grounded theory (GT) is an interpretive qualitative research method

which was originally conceived by Glaser and Strauss (1967). Like many

qualitative methodologies, GT is used in the collection and analyses of data.

However, while traditional research designs usually rely on a literature review,

leading to the formation of a hypothesis, followed by hypothesis testing, GT

attempts to investigate the phenomena of the real world through data analyses

without a preconceived hypothesis (Glaser & Strauss, 1967). In other words,

rather than beginning with a hypothesis or theory, grounded theory is a useful

methodology that fosters the development of theory through research activities

(e.g., observation, conversation, interviewing). Due to this inductive tenet, MT is

often described as being an emergent methodology (Glaser, 1992).

To better understand what grounded theory is, and what it is not, it is

useful to begin with some basic points. The following key characteristics of GT

are paraphrased from Creswell’s (1998) Qualitative Inquiry and Research

Design: Choosing Among Five Traditions. In this work, Creswell proposes that:

- the basic aim of grounded theory is to discover theory or theories.

- grounded theory relies on the premise that theoretical assumptions are

set aside by the researcher(s) prior to a study.

- the development of theory focuses on participant interaction in relation to

the phenomena under study.

27

- theory is derived from the analyses of data which are gathered through

field observations, interviews/conversations and documents (the literature

itself is treated as data).

- data analysis consists of systematic processes which begin as soon as

data is available.

- initial data analysis begins with the identification of categories, and the

connecting and comparison to other existing and emerging categories

(constant comparative method).

- the collection of subsequent data (also known as sampling) is based on

the emerging concepts generated by the constant comparison with newer

data.

- the standard format for data analyses includes three stages: open coding

(identifying categories, properties and dimensions), axial coding

(examining causal conditions, strategies, context and consequences) and

selective coding (development of propositions that usually follow a specific

“story line”).

- the resulting theory can be reported in either a narrative framework,

and/or as a set of theoretical propositions.

While this list can be extended and given greater detail, it provides a sketch of

the grounded theory methodology from initialization to completion. More detail

regarding grounded theory will be developed in the following sections.

28

Doing Grounded Theory

In the previous section, I have summarized some of the basic tenets of

grounded theory. However, since the seminal work, The Discovery of Grounded

Theory (1967), was written, there have been many interpretations of grounded

theory in both theory and practice. In fact, even the originators, Glaser and

Strauss, independently developed significantly different versions of GT

throughout their academic careers. While the variations in the perception of GT

are important23, there are two points that will help me push past these

differences, for now. First, as Dey (1999) puts bluntly, “If Glaser and Strauss

cannot agree on a definitive version, if would be impertinent (and indeed,

impossible) for an outsider to arbitrate” (vx). A dissertation could be written on

the incommensurability of grounded theory variations since the time of its

‘discovery’, and yet there is still strong support for the use of GT in academic

writing, including GT in dissertation research.24 My goal, in this case, is to tool

my use of GT from both a widely accepted form, and one that will enable the

discovery of new theory. Additionally, researchers must see methodologies as

enabling, and being a qualitative researcher, I understand that no amount of

methodology tinkering will produce research free of error. In other words, there is

as likely a chance to develop a ‘perfect’ form of grounded theory, as there is the

likelihood for this ‘perfect’ form to be accepted by all GT theorists. Yet, research,

mistakes and all, must go on. And I am comforted by Lincoln and Denzin (1994),

as they state “These criticisms and exchanges can operate at a level of

23 These will be discussed later in this paper. 24 The Grounded Theory Institute - http://www.groundedtheory.com/

29

abstraction that does little to help people who just go out and do research.” (p.

577)

In my construction of grounded theory in practice, I have broken down

the process of “doing” grounded theory research into three analytic phases.25

These phases include research design, data collection and analysis and

literature comparison. Within each phase, a series of plausible steps are

included. To better understand the applicability of grounded theory to my

intended study, I have used the context of studying the open source movement.

Yet, as this is a pre-proposal work, I have not yet gone into great detail.

As GT is both systematic and emergent, these steps are not intended to

be interpreted as strictly sequential. The linear focus, however, does help to build

a framework for my understanding of GT in practice, and will help in framing my

research proposal. Additionally, what is not made explicit within these steps is

the actual writing process. As GT uses constant comparison, writing is important

throughout the data analyses and comparison process. Writing happens from the

moment you begin data collection, through theory development and into later

literature comparative stages. Writing is at the heart of grounded theory.

The brief framework for “doing grounded theory” follows.

Phase 1: Research Design Phase

1) Review of technical literature

The open source movement is complex and bridges several disciplines

(e.g., philosophy, economics, sociology, social computing). While GT literature

25 These analytic phases are derived from the works of Glaser & Strauss (1967) and Strauss & Corbin (1997).

30

advises against an exhaustive approach to literature review (Glaser, 1992;

Glaser & Strauss, 1967; Strauss & Corbin, 1997), in the case of open source

(and certainly in the case of other complex multi-disciplinary topics), a strong

understanding of the basic technical literature is important, especially for aid in

the discovery of relevant categories26, relationships between categories and the

formulation of research questions. More specifically, Dey (1999) advises that

technical literature in grounded theory can be helpful in several ways:

a) for the development of knowledge of philosophical writings and existing theories b) as a useful bridge to secondary sources of data c) to stimulate research questions d) to foster the direction of theoretical sampling e) as a supplementary validation (an activity to occur after theory has been generated from GT analysis). (pp. 51-52)

2) Definition of Research Questions

As it is important in GT for theory to emerge from the data, the design of

the research questions is an important consideration. Generally, research

questions must be open and non-specific rather than formed as a specific

hypothesis. GT is not based on hypothesis testing. Questions formulated through

GT methods must enable theory generation, and not be restricted by

assumptions within the questions themselves. For instance, a possible question

for my study reads,

“Are there common values and beliefs held by members of open source communities, and if so, what are they?”

rather than 26 A category is considered to “stand by itself” as a “conceptual element of a theory”. (Glaser & Strauss, 1967)

31

“What common values and beliefs do members of open source communities share?”

The second question assumes that there are in fact shared beliefs to be found.

While this is a simplistic example, it’s important to develop questions that reflect

a basic understanding of the context of study, but will not make assumptions that

can hinder the emergence of theory from that data.

3) Sampling

Sampling, in the case of this potential study, can be described as process

of selecting a number of informants/participants from a defined study population.

The most common types of sampling in GT are known as purposeful (purposive,

selective) sampling and theoretical sampling. While in the literature, these terms

are often used interchangeably (i.e., selective sampling is the same as

purposeful sampling), some theorists make an important distinction.

Selective sampling may be seen to mean purposeful sampling …. Theoretical sampling may be seen as a variation of purposeful sampling, but purposeful sampling is not all necessarily theoretical sampling. (Coyne, 1997)

The distinction in the terminology of these sampling methods may seem

unimportant. However, in my interpretation of sampling as it applies to grounded

research, I see two distinct phases. When beginning the research, I plan to use

purposeful sampling to initially target specific members of the open source

community. In understanding the goal of selective sampling, I look to Patton

(1990; 2002).

32

The logic and power of purposeful sampling lies in selecting information-rich cases for study in depth. Information rich cases are those from which one can learn a great deal about issues of central importance to the purpose of the research.

As the sample begins to inform the development of propositions and theory,

further sampling should take on a responsive approach. This is where theoretical

sampling becomes important, as it is guided by the on-going analysis.

Theoretical sampling involves purposeful selection of samples to inform the emerging theory in the study. (Coyne, 1997) … conceptual elaborations . . . become the hypotheses which guide the researcher back to locations and comparative groups in the field to discover more ideas and connections from the data. (Glaser, 1978)

Where selective sampling leaves off, theoretical sampling can help to better

focus the efforts of the data collection on theoretically useful cases where the

researcher can test and/or extend theory.

Phase 2: Data Collection and Analysis Phase

4) Develop Varied Data Collection Sources

In regards to data collection, grounded theory studies share some

characteristics with other qualitative approaches to research. In particular,

sources of data are typically the same and can include interviews, field

observations and documents of all kinds (diaries, letters, autobiographies,

biographies, historical accounts, newspapers, and other media materials)

(Denzin & Lincoln, 1998). And in the case of a study related to the open source

movement, data sources would likely include electronic media, especially

33

listservs/majordomos, mailing lists, web logs, email transcripts and possibly

audio/video conference data.

While single sources of data are possibly appropriate with some

methodologies, the GT approach advocates the use of multiple data sources

which converge on the same phenomenon. Glaser and Strauss (1967) define

these as “slices of data” and write,

… no one kind of data on a category, nor technique for data collection is necessarily appropriate. Different kinds of data give the analyst different views or vantage points from which to understand a category and to develop its properties; these different views we have called slices of data. While the [researcher] may use one technique of data collection primarily, theoretical sampling for saturation27 of a category allows a multifaceted investigation, in which there are no limits to the techniques of data collection, the way they are used, or the types of data acquired. (Glaser & Strauss, 1967, p. 65)

What Glaser and Strauss (1967) are basically referring to in the use of multiple

sources of data is the concept of data triangulation. While there are various types

of triangulation28, data triangulation is where data are collected at multiple times

and/or from multiple sources to overcome problems of construct validity29 and

bias.

5) Data Collection Constant comparison is at the heart of the GT process. Therefore the

process of data collection must involve constant analysis of data as collected. In

27 Saturation, also known as theoretical saturation, is the point where existing and new data no longer bring about new theory or speculation, but continue to confirm theories already derived from data. 28 Some other common types of triangulation include investigator triangulation, methodological triangulation and interdisciplinary triangulation. 29 Construct validity refers the degree to which the testing/measurement tool used in a study accurately reflects the conceptual question of interest.

34

a grounded theory study, data collection and analysis works through the

following, mostly-overlapping phases.

Data Collection: As stated earlier, data collection sources vary. However,

interviewing is one of the most common collection methods. There are various

books related to good interviewing techniques and qualitative methods. Excellent

choices include Kvale (1996), Rubin & Rubin (1995), Seidman (1998) and

Wengraf (2001). As other sources of data are discovered, and subsequently

mined, relevant guidebooks will be consulted in considering appropriate methods

for data handling.

Note-taking: In GT, note taking is an important consideration in data

collection. In qualitative research, there are various approaches to the process of

note-taking, however Glaser & Strauss (1967) suggest a very specific approach.

The authors advise against audio recording or the production of word-by-word

transcripts. It is suggested that time is better spent focusing on additional

interviews than in producing transcripts or re-listening to recordings of interviews.

In GT, I understand that note-taking is meant to record basic concepts, rather

than full literal readings. However, I think a compromise can be argued here

where interviews can be recorded in order to check for understanding of the

basic derived concepts against the audio taped sessions. This slightly alternative

approach is supported by Dick (2002, Online).

Coding: There are three specific types of coding methods described in

typical ground theory literature. These distinct but similar methods of coding

include open coding, axial coding and selective coding. Open coding involves the

35

process of developing categories, concepts or themes which emerge from the

data. It is ‘open’ in the sense that the researcher engages in exploration of the

data without making prior assumptions about what might be discovered.30 Axial

coding is a coding method which serves to facilitate the building of connections

within categories and subcategories. Thus, this coding technique “serves to

deepen the theoretical framework underpinning your analysis” (Kerlin, 2002). In

selective coding, the researcher uses the categories identified through axial

coding to develop a “story line”. In this phase, conditional propositions (or

hypotheses) are typically presented” (Creswell, 1998, p. 57).

Memoing: An important activity of the GT methodology is the writing of

memos (memoing). Memos occur as a parallel process to data collection, note-

taking and coding. In effect, a memo is a note to oneself related to a possible

hypothesis or relationship between derived categories. Glaser (1978) developed

a set of “theoretical memoing rules” (pp. 89-91) which are useful in the

understanding the practical definition and boundaries of a memo. Additionally,

Corbin and Strauss (1990) emphasize the importance and place of memoing, as

stated in the following passage:

Writing theoretical memos is an integral part of the doing grounded theory. Since the analyst cannot readily keep track of all the categories, properties, hypotheses, and generative questions that evolve from the analytical process, there must be a system for doing so. The use of memos constitutes such a system. Memos are not simply “ideas”. They are involved in the formulation and revision of theory during the research process. (p.10)

30 Although the tabula rasa approach is an important principle of GT, it is one met with critique. This topic will be specifically discussed later in this paper.

36

Memoing is indeed an important component of data analysis in the GT

methodology. However, as you will see in the next section, the data analysis

process can now be assisted through the use of specialized computing software.

6) Using Qualitative Data Analysis Software

There are several qualitative data analysis software packages available

today (e.g., Atlas.ti, Nud*ist, Nvivo, etc). For this study, I will be using Atlas.ti

with my data analysis for three very practical reasons. First, the design of Atlas.ti

is based on grounded theory. The software is specifically helpful in the coding

and memoing process, and ultimately in helping to draw relationships between

categories and the developing theory. Second, Atlas.ti has a strong user

community.31 The community has helped to shape the functionality of Atlas.ti

over several versions, and is supportive of users of the software. And third, and

perhaps most important, I have access to local users of Atlas.ti who are willing to

help me through the process of data analysis.

In the past, the most common tools used in the data analysis process of

grounded theory included piles of divided blank paper, 5X7 memo cards, scissors

and possibly a photocopier. However, qualitative analysis software programs

have emerged to simplify and quicken the mechanical aspects of data analysis.

For some researchers, the greatest advantage of these programs is that it can

give the researcher more time to concentrate on the more creative aspects of

theory building. Tesch (1991) writes,

31 You can enter and join the Atlas.ti user community here - http://www.atlasti.com/maillist.shtml

37

The thinking, judging, deciding, interpreting, etc., are still done by the researcher. The computer does not make conceptual decisions, such as which words or themes are important to focus on, or which analytical step to take next. These analytical tasks are still left entirely to the researcher. (pp. 25-26)

And Lee and Fielding (1991) also support my original idea with,

It is likely that computers will bring real benefits to qualitative researchers, making their work easier, more productive and potentially more thorough. (p. 6)

There are two principal levels (modes) available when working through

Atlas.ti: the textual level and the conceptual level. With the textual level, the

researcher focuses on breaking down or segmenting the primary documents (i.e.,

data files) and as well, focuses on the writing of memos. In essence, the

different types of GT coding (i.e., open, axial, selective) are directed by the types

of memoing which are done in Atlas.ti. For instance, “code memos” relate to

open coding and focus on conceptual labelling. Additionally, “theoretical memos”

in Atlas.ti relate to axial and selective coding. Then, in Atlas.ti’s conceptual level,

the researcher is able to visually connect selected portions of the “data, memos

and codes into diagrams which graphically outline complex relations” (Atlas.ti,

1997, p. 7). The conceptual level allows the researcher to examine and explore

relationships between data that may not have been as obvious as in the textual

view.

I can go on with a description of the step-by-step process of using Atlas.ti.,

however, that information is likely more appropriate in a technical manual. My

purpose in this short explanation was to put forth how qualitative analysis

software packages can be appropriate in grounded theory, and valuable to the

38

researcher and the research. Further information regarding Atlas.ti can be found

at http://www.atlasti.com.

Phase 3: Literature Comparison Phase

7) Comparison With Extant and Emerging32 Literature

In using a software package such as Atlas.ti, the goal of the data analysis

(in a grounded theory study) is to reach a point of “theoretical saturation”.

Theoretical saturation “refers to the (non)emergence of new properties,

categories, or relationships. Once the data no longer offer any new distinctions of

conceptual import, categories can be described as ‘saturated’ and no further

evidence need to be collected” (Dey, 1999, p. 8). In GT, at this point, the results

of the emerging theory are often written up in the form of a narrative, presented

and the research could be regarded as complete. (Glaser & Strauss, 1967)

However, to strengthen both the level of analysis and the theoretical findings, it is

wise to continue the constant comparative process and once again, explore and

examine the derived theories with relevant literature. Eisenhardt (1989) writes,

Overall, tying the emergent theory to existing literature enhances the internal validity, generalisability, and theoretical level of the theory building from case study research … because the findings often rest on a very limited number of cases. (p. 545)

When I first started contemplating this research in the open source

movement, the literature of the movement was very much at the fringe. While

there was much written about the open source movement as a general

32 I use the term ‘emerging’ in a fairly specific sense. Much of the literature found regarding open source is either unpublished (in the traditional sense), or is constantly being modified (through collaborative licenses such as found through the Creative Commons).

39

philosophy (Stallman, Raymond), there was little written on how we could learn

about the open source movement as a structure for collaborative practice in

fields outside of computing science (e.g., programming). However, recent works

have begun to extend the idea of the concept of open source collaborative

methods in other disciplines. A few examples, with brief descriptions, include:

The Business and Economics of Linux and Open Source (Fink, 2003) – With open source software being “given away”, corporations are trying to better understand the emerging business model that has made companies such as RedHat, Suse Linux and Novell recently more successful. Fink’s book is useful as a manager’s guide to understanding and embracing the open source movement in the corporate setting. It focuses on understanding the open source model’s usefulness in cost-cutting, rediscovering corporate “value” and implementing collaborative techniques across and beyond the corporate enterprise. The Success of Open Source (Weber, 2004) – Her, Weber traces the philosophical and technical foundations of the open source movement and comes to the conclusion that society’s foundational assumption of “knowledge as property” has irrevocably shifted. As professionals and citizens of the knowledge economy, we are left to reinvent and arrange our world of information on this newly accepted foundation. Free Culture: How Big Media Uses the Technology and the Law to Lock Down Culture And Control Creativity (Lessig, 2004) – While this book is not entirely related to the open source movement, it addresses an important common theme: knowledge as property. Lessig’s position on the topic is fairly evident (look no further than the ad hominem title), and the book is significant not only in the content it describes, but through the way in which it was released (i.e., free through a Creative Commons license AND for retail sale through Amazon.com) as well as the movement it provoked (e.g., freeculture.org developed by the students of Swarthmore College).

While much of the literature seems to present certain general assumptions

(e.g., collaboration amongst individuals is constructive, the open source model

represents an important model for understanding collaboration), it is important for

40

this stage in GT to seek and examine conflicting literature. Eisenhardt (1989)

writes,

By examining conflicting literature and seeking explanations for why there are contradictions one can preempt criticisms and enhance confidence in the findings, but looking at conflicting theory also represent an opportunity for deeper insights. By looking at literature with similar findings, one connects the current study to previous work, corroborate the findings and strengthen confidence. (p. 544)

Through thorough data analysis and the examination of a wide range of related

literature, theoretical constructions are perceived to be strengthened and better

“grounded”.

8) Producing a Final33 Narrative

The concluding phase of a grounded theory study may not look

significantly different than the final chapter of a study done using ethnography,

action research or an other qualitative methodologies. However, what may be

distinct in some cases is that the focus for grounded theory, throughout the

process, remains focused on the development of theory. While theory can be

expressed in various ways (e.g., organizational charts, conceptual diagrams, flow

charts, etc.), many GT researchers find the use of a narrative helpful in

summarizing and describing theoretical propositions.

Grounded theory has been described as a “mixed marriage” and a

reconciliation of naturalistic inquiry with “the rigorous canons of variable analysis”

(Dey, 1999, p. 45). Glaser & Strauss (1967) envisioned GT in this way in order

33 Finality is a concept that can suggest a rigid epistemological approach to a narrative being ‘the truth’ or ‘the final word’. From my stance, I refer to the term ‘final’ to suggest that this work, as in the case of academic ‘for credit’ work, must face the reality of having a final ‘condition’ as to which it is assessed.

41

to provide a “confidence in the data” (p. 68) as which to base theory. While much

of the described processes in the previous section (note-taking, coding,

memoing, etc.) seem quite prescriptive, there is little in the originating literature to

describe the specific actualization of processes (e.g., memoing), and therefore,

the method is open to sufficient interpretation.

A straight-forward and prescriptive outline for the “findings” section of a

GT study is described by Creswell (1998):

The findings section presents the theoretical scheme. The writer includes references from the literature to show outside support for the theoretical models. Also, segments of actual data in the form of vignettes and quotes provide useful explanatory material. This material helps the reader form a judgement about how well the theory is grounded in the data. (p. 179)

This format is common, and for most readers, what Creswell describes is not

specific to grounded theory studies only. However, although being somewhat

generic, this format does suit grounded theory research well, especially in its

positioning of the data and literature (as data) throughout the body of the

narrative.

A point of contention with grounded theory emanates from the guiding

principle which can be summarized as “theories are revealed rather than

contrived” (Dey, 1999, p. 35). While this is an important principle and useful in

describing the inductive foundation of GT, it seems to give greater relevance to

the act of discovery over that of creativity. In thinking of how I will conclude my

study with reverent heed to my understanding of GT, I have realized that my

current thinking will involve both creativity and discovery especially in the latter

42

descriptive parts of my research. Let’s consider these two terms with appropriate

synonyms.

Discover: produce insight, expose to view, become aware, disclosure, manifest, to make known (for the first time) Construct: compose, draw, frame, build, combine, fit together, delineate, assemble

It should be fairly obvious that while discovery is often seen as a preferred

mindset in GT research, a thoughtful balance of both discovery and creativity will

be essential in producing a document that moves beyond a strict summary of

findings, implications and recommendations.

Considering Other Qualitative Methods

Qualitative inquiry or qualitative research is used to describe various

orientations to interpretivist research. There are many accepted forms of

qualitative inquiry. Some of the most commonly used include ethnography,

narrative inquiry, phenomenology, action research, case study34 and of course,

grounded theory (Creswell, 1998; Denzin & Lincoln, 1994; Glesne, 1999; Patton,

1990). While choosing a specific methodology may seem a bit daunting,

understanding the appropriateness35 of a particular approach in relation to “what

you want to understand” is a helpful place to begin.

34 There is some contention whether case study is a ‘true’ methodology or simply a method to be used within other frameworks. 35 I use the term appropriateness instead of inferior (as posed in the original comprehensive question) as I believe that important question to be asked by a researcher is what methodology is most appropriate for understanding and answering the research questions I have asked. In being an qualitative researcher, the term ‘inferior’ is subject to interpretation and debate. That is another paper altogether.

43

Tesch (1990) developed a useful framework for the categorization of

several categories of qualitative data. Using this framework, Tesch (1990)

describes the qualitative methodologies that are likely most appropriate for

working with the identified data category. The four categories of data with the

appropriate methodological approaches include: a) characteristics of language

(discourse analysis, ethnoscience, symbolic interactionism); b) discovery of

regularities (grounded theory, critical research, ethnography); c) discerning

meaning (phenomenology, case study, hermeneutics); and d) reflection (action

research, educational connoisseurship, reflective phenomenology, reflective

autobiography, heuristic research).

In thinking about the research questions that guide my study36, it is simple

to reject three categories immediately. For instance, I am not specifically

interested in language or the characteristics of language, and therefore will not

pursue methodological approaches such as discourse analysis Although in

saying this, I would be hard-pressed to discount my critical bias toward

acknowledging the power and meaning of and within discourse. I have no doubts

that indeed it would be interesting to analyze the discourse of the open source

community. This is a community, in my perception, which has emerged (in part)

in reaction to an imbalance of power in the world of information technology and

knowledge. It would be interesting to analyze the discourse of the community

from its foundational moments as a fringe movement, up unto its new reality as a

proximal mainstream alternative to proprietary software giants. With no doubt,

taking this route would lead to an important and revealing study. However, such 36 These questions were made available to my committee in the pre-comprehensive paper.

44

questions require a different methodology, and these are not the questions I

seek. For now, I am happy to put Norman Fairclough and Michel Foucault back

on the shelf.

I can also disregard the category reflection for the moment. One of

methodologies that did appeal to me when considering this research was that of

action research. I have had some experience with action research, and I know

that there is strong support in the Faculty of Education in this area. However, the

question of appropriateness arises again when I ask myself what I want to learn.

To explain the inappropriateness of action research in this study, I will be begin

with a few representative definitions for action research.

… systematic enquiry designed to yield practical results capable of improving a specific aspect of practice and made public to enable scrutiny and testing. http://www.standards.dfes.gov.uk/research/glossary Inquiry-based research conducted by teachers that follows a process of examining existing practices, implementing new practices, and evaluating the results, leading to an improvement cycle that benefits both students and teachers. http://cs3.wnmu.edu/elearning/a404/support/a404b0_50100.html A (usually cyclic) process by which change and understanding can be pursued at the one time, with action and critical reflection taking place in turn. The reflection is used to review the previous action and plan the next one. http://www.uq.net.au/action_research/arp/actlearn.html

In looking at the various definitions of action research37, it was quite

obvious that the reasons for action research were not compatible with my

intended research. In studying open source, I am not looking to improve practice

37 Many more definitions of action research provided by Google at: http://www.google.com/search?sourceid=navclient&ie=UTF-8&oe=UTF-8&q=define%3A+action+research

45

or produce action in the world of my research participants. Yet, in saying this, I

do see the potential for action research in the area of open source. For instance,

one could study a local open source advocacy group such as LOSURS

(Linux/Open Source Users of Regina Saskatchewan) and study their practice.

This particular group focuses on bringing awareness of open source software to

schools, government, business and individuals through volunteer-organized

conferences, installfests and other events. It would be interesting to study the

activities of this group, and focus on their reasoning and the improvement of their

practice. This is just one example of many possibilities, however, such research

activities do not fall into the frame of my intended study.

Lastly, looking again at Tesch’s (1990) framework, I have also made the

decision that my study will not focus on discerning meaning. Phenomenology is

a major approach in helping researchers recognize and comprehend meaning of

the lived experiences of individuals in relation to particular phenomena (Creswell,

1998). Phenomenology is based on the works of Edmund Husserl and Alfred

Schutz (Denzin & Lincoln, 1998), and is both a philosophical and psychological

approach to research. Bowling (1997) describes phenomenology as “The

philosophical belief that, unlike matter, humans have a consciousness. They

interpret and experience the world in terms of meanings and actively construct an

individual social reality” (p. 31). Of course, a phenomenological study could

easily be structured in an open source community, focusing specifically in how

members of the community make meaning with their world, and with each other.

46

However, looking back at my intended questions, it should be easy to see that

this is not the intent of my study.

So Why Grounded Theory?

There are several reasons that I have chosen grounded theory over the

many other possibilities and choices in methodological approaches. Here is a

brief overview of the advantages of grounded theory in regards to this proposed

study.

1) Focus on theory development

When Tesch (1990) categorizes grounded theory as a methodology that

assists in the discovery of regularities, she is referring partially to the

methodology’s focus on theory development. Theory has always been an

important part of human contemplation. From Aristotelian epistemology, to

modern scientific and social research, the attempt to develop explanations has

proven extremely useful. In the pragmatic tradition, I see theory especially useful

in the following ways.

Integration and Explanation of Knowledge: There is much known and

unknown about the open source movement. However, it has undeniably become

an important model for contemplation (Fink, 2003; Lessig, 2004; Weber, 2004).

As one of the very few educational researchers studying the open source

movement in relation to teachers, it would be useful to integrate as much existing

knowledge in this area with the emerging knowledge to be gained from this

study. The grounded theory methodology provides a systematic approach to the

47

building of relationships and theory, and possibly explaining the open source

movement through a particular lens.

Predicting What Is Not Yet Known/Learned: Theory developed through

grounded theory can help us understand things as they exist, and as well, help

us project how things may someday be. Hargreaves (2003), in touting the open

source philosophy, writes, “A key to the transformation is for the teaching

profession to establish networks the capture the spirit and culture of hackers –

the passion, the can-do, the collective sharing” (Online). While I really like this

speculative comment from Hargreaves, and in some ways, agree with it (or more

so, want to agree with it), there is little in his report that supports his statement.

However, grounded theory helps to establish a systematic, yet emergent

methodology which can allow researchers to produce supported theory.

Improving the Human Condition: I would not partake in social research

unless I felt that the results of the research are beneficial to individuals and

society. I feel that better in understanding the open source movement, we may

improve our educational system, develop a better understanding of human

collaboration, provide people with greater access to information and technology

and ultimately improve our lives. In my opinion, theory is most valuable when we

attempt to apply and advance existing knowledge in order to solve our basic

human problems.

48

2) GT provides guidelines and ‘rules of thumb’

Grounded theory should not be viewed as a strictly prescriptive and rigid

approach to research. In reality, GT is simply a set of guidelines or ‘rules of

thumb’ which can facilitate data collection and analysis. Grounded theory

researchers are encouraged to develop their own research style which refers

both to the researcher’s personal abilities, as well as the context/content of the

research in question (Strauss, 1987).

While the procedures are designed to give the analytic process precision and rigor, creativity is also an important element. For it is the latter that enables the researcher to ask pertinent questions of the data and to make the kind of comparisons that elicit from the data new insights into phenomenon and novel theoretical formulations. (Strauss & Corbin, 1990, p. 31)

While there are other methodologies that support a less rigid process for

data analysis, this methodological approach is appealing (to me) as it

provides both structure and flexibility.

3) Interest in Using Qualitative Analysis Software

A simple, yet honest reason for the use of grounded theory is that this

methodology is best suited for a qualitative analysis package such as Atlas.ti.

Alone, this would not be enough of a reason to choose this specific methodology,

but as computing is a large part of my life, and an avid interest, I look forward to

better understanding how software can be helpful in specific types of research.

My comfort zone often exists behind the keyboard, and computing certainly plays

a large part in my way of knowing the world. Thinking about the long journey

ahead of me, I need to be comfortable with the way I will conduct this research

49

and in how I will handle the tremendous amount of potential data. For now, it’s

jacket and shoes off, and let the glowing screen be my companion.

Conclusion

In the preceding paper, I have discussed the nature of grounded theory,

practical reasons for to using this approach and motivations for choosing this

methodology. As a fairly novice researcher, I am looking forward to learning and

applying this approach to my research questions. I am not convinced that I will

stay in this ‘zone’ for the rest of my research career, but at the very least, I hope

to learn valuable things with this particular approach to research. Although there

are so many choices of research methodology, I am comfortable with this

position for my the duration of my dissertation. I believe that grounded theory will

allow me to explore and develop the inquiries that I pursue.

50

Question #3:

3) Explain change theory, innovation theory, communities of practice

theory and social capital theory. Build/create/describe a conceptual

framework that will guide your investigation of the open source

movement in education.

51

Introduction

For this final paper, I will discuss three major areas of study. The topics

include change theory (CT), communities of practice and social capital theory.

To be direct, these are very large and complex areas of study, and I have limited

space to put forth my consideration of this content. Even in studying just one of

these topics, the twenty-page space could be seen as inadequate to present a

comprehensive understanding. Therefore, in this section, I will attempt only to

present a thumbnail sketch of some of the basic, relevant information pertaining

to each field. If this seems inadequate, I point your attention to three more

comprehensive papers that I have written previously in these areas. The papers,

titled below, can be found at the following URLs:

- Innovation, Change Theory and the Acceptance of New Technologies: A Literature Review http://educationaltechnology.ca/couros/publications/unpublishedpapers/change_theory.pdf - Communities of Practice: Literature Review http://educationaltechnology.ca/couros/publications/unpublishedpapers/communities_practice.pdf - Open Source Communities of Practice and the Introduction of Technological Innovation http://educationaltechnology.ca/couros/publications/unpublishedpapers/opensource.pdf

Following the discussion of these areas of study, I will attempt to build a

conceptual framework in relation to my prospective study (the open source

movement). However, in considering this, I will bring one possible concern to the

attention of my supervising committee. In understanding one of the key guiding

principles of grounded theory (as highlighted earlier), it is important to consider

the place of literature in the process of induction-based research. As explained

earlier, Glaser (1967) strongly warned against preconceived theories as working

against and contradicting the induction process. If I work with a particular theory

52

in mind, and try to place the data within that theory, this has now become

deduction, and it is no longer, in the strictest sense, grounded theory. Therefore,

the framework I build here should only be viewed as a referential one which can

be consulted as data and theory emerge. The framework itself, should not be a

key guiding force of my study, but rather a frame of reference which can either be

regarded as appropriate and applicable, modified as to fit the emerging theory, or

simply discarded.

Introduction to Change Theory

“Change theory” is really an umbrella term which envelops numerous sub-

theories that describe change within various contexts. In education, CT is

perhaps most frequently used in the areas of leadership, administration and

educational technology. Change theorists, in general, attempt to understand the

change process and provide and build strategies to effect change (Ellsworth,

2000; Fullan, 2001). I believe that change literature will prove to be important in

understanding the open source movement. However, in qualifying this belief, it is

important to understand and describe the various assumptions underlying much

of the popular change literature. These fundamental assumptions should be

acknowledged when one considers applying CT to a particular context (e.g., the

open source movement).

53

Educational change theory can be seen to be influenced by the following

key assumptions:

1) Change can be understood and managed. The change literature that I have studied and will describe in this paper is built on the idea that change can be planned, initiated and developed through leadership or administrative strategies. In this sense, the focus of much change theory is often pragmatic; change is studied so that it can be managed. (Ellsworth, 2000; Rogers, 1995)

2) Planned change is focused upon introducing innovation or

innovations to individuals or within a system. Planned change may be limited to implementing a particular technology into an instructor’s classroom (e.g., use of PowerPoint or a data projector as an instructional tool), or planned change can be focused toward more encompassing wide-scale school reform projects. (Bates, 2000; Ellsworth, 2000; Fullan, 2001; Fullan & Miles, 1992; Rogers, 1995)

3) Planned change is value-laden. When introducing change to an

individual, group or system, at some point leadership38 has made a decision as to a particular innovation being valuable, or more valuable than items or processes which currently exist. And within the change process, value is also negotiated as the individuals closely affected by the change process may be accepting, neutral or resistant to the innovation for reasons related to perceived value. Thus, leadership of change often involves the engagement of different sets of interests, interpretations and identities, which may allow the mutual adaptation and ‘success’ of a proposed change project. (Fullan, 2001; Gordon, 2002; Knight & Trowler, 2001; Ramsden, 1998; Zaltman & Duncan, 1977)

4) Planned change requires people. There is little debate that people are

at the heart of the change process. And, in much of the classic literature the term change agent is used to describe the person (or persons) who is the initiator of the change effort. Change agents may be identified as officials hired by an institution to be responsible for particular changes (e.g., a school division’s technology consultant), or more informally, a change agent may simply be an influential person who causes change at a micro-level (e.g., the tech-savvy teacher next-door). (Fullan, 2001; Gladwell, 2000; Hall & Hord, 1987; Rogers, 1995)

5) Planned change is complex and often requires multiple approaches

and tools. Theorists have gone to great lengths to develop models and checklists for change. While some attention has been given to the

38 Leadership is meant as a general term that can mean anything from an individual leader, to an entire administration.

54

individual or potential adopters (Hall & Hord, 1987), other theorists have developed broad strategies for promoting change within larger environments. (Bates, 2000; Fullan, 2001; Havelock & Zlotolow, 1995) Change is complex and even with the combination of such strategies, planned change is not always achieved.

Diffusion of Innovations Theory

In trying to comprehend change theory, it is useful to trace CT research

back to its modern historical roots. Change theory, as it exists today, has been

strongly influenced by diffusion of innovation theory (DIT) which first appeared at

the beginning of the 20th century. Understanding the philosophical foundations of

change theory is an important step in understanding current change theory.

Ralph Waldo Emerson is reported to have once said, “Make a better

mousetrap, and the world will beat a path to our door.” This principle that may

have held merit in the early 1800’s, might easily be dismissed in the post-

industrial world. Theoretically superior technologies such as Betamax recorders,

the Dvorak keyboard and the early Apple operating system have succumbed to

the VHS standard, QWERTY and the Wintel39 monopoly, respectively. It is in

general agreement that the adoption of technology is a more complex process

than the technical superiority of a product (Abrahamson & Rosenkopf, 1997;

Reigeluth & Garfinkle, 1994; Rogers, 1995; Ryan & Gross, 1943).

Diffusion of innovations theory provides a useful (albeit incomplete) lens

for assisting researchers in understanding the complexity related to the

acceptance or rejection of innovation. In its basic form, diffusion is defined as the

process by which an innovation is adopted and gains acceptance by individuals

39 Wintel is the common trade term used to describe personal computers based on the Intel architecture and the Windows Operating system. This has by far become the prevalent configuration for standard personal computers.

55

or members of a community. DIT comprises several sub-theories that

collectively study the processes of adoption. The first famous account of DIT

research was done in 1903 by French sociologist Gabriel Tarde (1903). Tarde

plotted the original S-shaped innovation curve (see Appendix A) as he believed

that most innovations have an S-shaped rate of adoption. Through the slope of

the S-curve, Tarde could identify those innovations with a relatively fast rate of

adoption (steep slope) versus those with a slower rate (gradual slope). Since

Tarde, the S-slope has become important for those studying the adoption of

ideas, especially those found in business.

Several decades later, Ryan and Gross (1943) published their seminal

study which described the diffusion of hybrid seed among a group of Iowa

farmers. At the time of the study, U.S. farms were slowly becoming business

enterprises rather than family subsistence units. As corporations entered into the

business of agriculture, so did the concerns of higher productivity, efficiency,

competitiveness and agricultural innovations. Ryan & Gross wanted to study the

process in which innovations in agriculture were adopted. They discovered that

diffusion was “a social process through which subjective evaluations of an

innovation spread from earlier to later adopters rather than one of rational,

economic decision making” (Valente, 1995). At the time, this was a novel

perspective on the diffusion process and emphasized the effect of social factors

on adoption.

Ryan & Gross (1943) also noted that the rate of adoption among those

studied followed an S-curve when plotted on a cumulative basis over time. This

56

supported the work of Tarde reported 40 years previously, and renewed interest

in Diffusion Theory. Additionally, Ryan and Gross (1943) classified their study

participants (Iowa farmers) into five adopter categories. These categories

included: innovators, early adopters, early majority, late majority and laggards.

Theorists since (Abrahamson & Rosenkopf, 1997; Gladwell, 1996; Midgley &

Dowling, 1978; Rogers & Kincaid, 1981) have used and modified these basic

categories to build upon this early work. What is also important from these

studies is the distinctive characteristics of each adopter level. For instance, Ryan

& Gross (1943) identified that those farmers most likely to adopt (innovator

category) were more ‘cosmopolite’ and belonged to a higher socioeconomic

status than members of the other categories (later adopters). While the work of

Ryan & Gross (1943) began the next wave of Diffusion research, the next

seminal work in the area would not appear until almost two decades later.

Everett Rogers and the Diffusion of Innovations

Everett Rogers’ claims that his 1995 text, Diffusion of Innovations, is a

synthesis of over 3800 DIT publications. While much of his theory originates in

rural sociology, his established framework has been used in diverse areas such

as business and marketing, anthropology, public health, and of course,

education. Rogers defines diffusion as “the process by which an innovation is

communicated through certain channels over time among members of a social

system” (1995, p. 5). Diffusion theory, in this light, is very much a

communications-theory based model. The process Rogers (1995) refers to is

mediated through the two-process of communication convergence (Rogers &

57

Kincaid, 1981), rather than a one-way linear act. Additionally, diffusion is a

special type of communication in which the messages pertain to a new idea.

This is important in that the diffusion process is inherently uncertain due to the

newness of the idea and as to how the message (diffusion) will be accepted.

Key to Rogers’ (1995) definition of diffusion is the presence of four

elements in the diffusion of innovation process. These elements include the

following:

1) The Innovation: an idea, practice(s) or objects that is perceived as new by individuals or a group of adopters.

2) Communication Channels: the means by which innovations move from

individual to individual, or group to group. 3) Time: the non-spatial interval through which the diffusion events occur.

These events include the innovation-decision process, the relative span of time for the individual or group to adopt the innovation and the innovations’ rate of adoption in a system.

4) A Social System: a set of interrelated units that are engaged in joint

problem solving activities to accomplish a goal or goals.

These components are a bit simplistic, however, I find this communications

theory-based model useful in helping to simplify some of the complexity the

diffusion process. As Rogers’ (1995) work represents a compilation of the

majority of the previous DIT research, this model could be helpful in studying

many types of technological innovation.

Rogers (1995) also goes further to identify important characteristics of

innovations as perceived by individuals. These are important as they are

constructed as to the way in which potential adopters may (sometimes

58

unconsciously) view the innovation. The characteristics, which forms the basis for

what is regarded as perceived attributes theory, include:

1) Relative advantage: the degree in which an advantage is perceived as better than the idea it supersedes.

2) Compatibility: the degree to which an innovation is perceived as being

consistent with the existing values, past experiences and needs of potential adopters.

3) Complexity: the degree to which an innovation is perceived as difficult to

understand and use. 4) Trialability: is the degree to which an innovation may be experimented

with on a limited basis. 5) Observability: the degree to which the results of an innovation are viable

to others. The easier it is for individuals to see the results of an innovation, the more likely they are to adopt it. (Rogers, 1995, pp. 15-16)

Although, I don’t feel that the adoption process is limited to these perceived

attributes, I feel that these elements would be helpful in formulating questions for

potential adopters in better understanding what factors make adoption possible

or desirable. Additionally, although Rogers brings up the idea of reinventing

innovation (e.g., an adopter adapting an innovation to a specific need), these

characteristics do not fully account for this process. The idea of reinvention and

what I would call personalization of innovation, especially in regards to a

teacher’s use of technology, would be an important feature for consideration. If I

were to consider these characteristics in future research, I would pay special

attention to the idea of reinvention as it seems to be an element missing in

Rogers’ research.

Rogers (1995) distinctly separates the diffusion process from the adoption

process. While the diffusion process permeates through society and groups, the

59

adoption process is most relevant to the individual. Rogers (1995) defines the

adoption process as “the mental process through which an individual passes

from first hearing about an innovation to final adoption” (p. 35). The five steps in

this process are regarded as 1) knowledge (awareness), 2) persuasion (interest),

3) decision (evaluation), 4) implementation (trial) and; 5) confirmation (adoption).

Throughout the adoption process, the individual seeks knowledge of and skills

which will ultimately affect the adoption process. For a potential adopter, the

process will proceed through the various steps and lead to adoption, or

alternately, lead to rejection of the innovation. (Rogers, 1995)

Rogers also offers a scientific approach to understanding the rate of

adoption. Rogers (1995) has developed five variables which affect the adoption

rate of any particular innovation. These include 1) perceived attributes of

innovations (discussed earlier), 2) type of innovation-decision, 3) communication

channels, 4) nature of the social system, and; 5) extent of change agents’

promotion efforts. Rogers’ model could help a researcher to consider the basic

forces which affect both adoption rates, and the factors which may lead to the

rejection of an innovation. However, in its own simplicity, which may be ironically

its strength, it is limited when explaining more complex human systems.

However, in qualifying this, a schematic description of this model is shown below

in Figure 2.

60

Figure 2: Variables Determining the Rate of Adoption of Innovations

(Rogers, 1995, p. 207)

Significant Change Theory Literature

While the work of Everett Rogers has been highly influential, there are

many other theorists that should be considered when understanding change

theory literature. Ellsworth (2000) has developed a useful concept map for

acknowledging the key areas in current change literature. The concept map is

meant to epitomize seven major models represented by various CT theorists.

See figure 3 below.

61

Figure 3: 360° View of the Educational Change Process

Model adapted by McGriff (2001, Online) As you can see, Rogers’ (1995) work , while important, represents only

one particular model in the wider frame of change theory. The following are

abbreviated descriptions of some of the change theory models identified by

Ellsworth (2000).

1) Fullan & Stiegelbauer (1991): The role of the change agent is stressed in this work as the authors explore the implications of educational change for people and organizations promoting change at various levels. Individuals and organizations influencing change include teachers, principals, students, district administrators, consultants, parents (communities), government and teacher-educators.

2) Havelock & Zlotolow (1995): In The Change Agent’s Guide, the authors focus on developing a checklist approach to prepare facilitators of change with appropriate tools and a strong understanding of how and when to use them. This work helps to bring about a practical understanding of a change process, similar to the one described by Fullan and Stiegelbauer (1991).

62

3) Hall & Hord (1987): The Concerns-Based Adoption Model (CBAM) was developed by Hall & Hord as a tool that could be used to facilitate change in organization. It touts a personalized, bottom-up approach to change and utilizes what is identified as the “seven stages of concern”. These stages include awareness, information, personal, management, consequence, collaboration and refocusing. The stages are meant to understand and facilitate an individuals movement through the change process. (See Appendix B for a visual representation of the stages of concern).

4) Zaltman & Duncan (1977): In Strategies for Planned Change, the

authors identify some of the most common reasons that individuals reject or accept change. In better understanding resistance to change, practitioners may be better able to address the underlying barriers to change, or perhaps create environments were these barriers are less frequent.

5) Ely (1990a; 1990b; 1976): While Zaltman & Duncan (1977) focused on

various barriers to change, Ely focuses more specifically on the environment and the context in which planned changed is implemented. For instance, Ely (1990a) suggests that dropping computers into the educational environment with nothing but its assumed value does not provide a favourable condition for change.

Final Thoughts on Education Change Theory

Educational change theory, within the most popular variations, focuses on

planned change. Change theory is especially helpful in understanding how

change, and more specifically, how the adoption of particular innovations can

best proceed within educational institutions. In the case of open source, in some

contexts wide-scale implementation is wholly planned. One can look to wide-

scale open source (inspired) implementations such as the thin-client installation

in the Northwest Catholic School Division40 or the provision of Star Office in all

Ontario K-12 schools41. However, in many cases, the use (and acceptance) of

40 http://www.canopener.ca/article.php?story=156 41 http://www.itbusiness.ca/index.asp?theaction=61&sid=55732

63

open source is not wholly planned, and rather a result of other factors. At this

point, planned change theory models may be of less value for analysis.

To close this section, I have included a quote from Gabriella Coleman

(1999), a theorist who has spent much time on considering the open source

movement. In describing the open source movement, and in considering the

many intangibles, she writes:

...the meanings, aims, visions, and aspirations of the open source community are difficult to pin down. Unlike the initial quote, which posits a unitary vision and goal for Linux hackers, closer inspection of the movement reveals a cacophony of voices and political positions: anarchic ideals of freedom, "tribal" gift-economy rhetoric, revolution, Star Wars imagery, web manifestos, evangelization to the corporate sector, the downfall of the "Evil Empire" (a.k.a. Microsoft), grass roots revolution, consumer choice and rights, community good, true market competition, DIY (Do it Yourself) culture, science as a public good, hacker cultural acceptance, functional superiority, and anti-Communist rhetoric are but a number of the terms, images, and visions promulgated by and attached to the open source community. (Online)

The open source movement is not merely about the implementation of new

technologies into a system. While change theory may be helpful, I feel there must

be a better understanding of the social systems and social influences

surrounding the communities in which open source philosophies exist.

Introduction to Communities of Practice Literature

While there are many definitions of communities of practice (CoPs), I find

the definition from Hildreth & Kimble (2000) relevant and consistently useful to

my particular interests. The authors define CoPs as “a group of professionals

informally bound to one another through exposure to a common class of

64

problems, common pursuit of solutions, and thereby themselves embodying a

store of knowledge” (p. 3). Another useful definition comes from Wenger,

McDermott & Snyder (2002) who define CoPs as “groups of people who share a

concern, a set of problems, or a passion about a topic, and who deepen their

knowledge and expertise in this area by interacting on an ongoing basis” (p. 7).

While the use of the term ‘communities of practice’ has become widespread, the

term actually stems from classic theories based on the idea of learning as social

participation (Wenger, 1998a; 1998b). To better understand the concept of

CoPs, it’s important to have a solid understanding of social learning theory. For

a brief overview of social learning theory, see Appendix C.

Etienne Wenger is one of the most prominent theorist in the areas of both

social learning theory and communities of practice. In Communities of Practice:

Learning, Meaning, and Identity, Wenger (1998a) posits that today’s modern

institutions are largely based on the assumption that “learning is an individual

process, that it has a beginning and an end, that it is best separated from the rest

of our activities, and that it is the result of teaching” (p. 3). Within the context of

social learning theory, the idea of learning in this sense is displaced. Learning

becomes, fundamentally, a social phenomenon and is placed in the context of

our lived experience and participation in the world (p. 3). In bringing forward his

ideas of social learning theory, Wenger starts with four main premises:

1) We are social beings. Far from being trivially true, this fact is a central aspect of learning.

2) Knowledge is a matter of competence with respect to valued

enterprises – such as singing in tune, discovering scientific

65

facts, fixing machines, writing poetry, being convivial, growing up as a boy or a girl, and so forth.

3) Knowing is a matter of participating in the pursuit of such

enterprises, that is, of active engagement in the world.

4) Meaning – our ability to experience the world and our engagement with it as meaningful – is ultimately what learning is to produce. (Wenger, 1998a, p. 4)

Here Wenger puts forward that learning is part of a more encompassing process

which places individuals as active participants in the practices of social

communities.

Wenger (1998) also presents components which he says are necessary to

characterize social participation as a process of learning. These include the

following:

1) Meaning: a way of talking about our (changing) ability – individually and collectively – to experience our life and the world as meaningful.

2) Practice: a way of talking about the shared historical and social

resources, frameworks, and perspectives that can sustain mutual engagement in action.

3) Community: a way of talking about the social configurations in

which our enterprises are defined as worth pursuing and our participation is recognizable as competence.

4) Identity: a way of talking about how learning changes who we

are and creates person histories of becoming in the context of our communities. (Wenger, 1998a, p.5)

These elements are deeply interconnected and mutually defining. Wenger

(1998) also provides a visual representation of his model, found below in Figure

4. When Wenger describes communities of practices, he intentionally positions

this concept within a larger conceptual framework. The four elements described

66

above (practice, community, identity, meaning) are important as they are

interchangeable with their relationship to learning. For instance, in Figure 4

below, you may switch any of the elements with learning, and the structure still

makes sense. Learning can be central or peripheral to the process, but remains

always an important component.

Figure 4: Components of social theory of learning: an initial inventory

from Wenger, 1998, p. 5

According to Wenger (1998), a community of practice defines itself along

three dimensions which are related to practice itself. The first component is

mutual engagement. Practice does not exist in the abstract, so CoPs reside

around people engaged in certain common actions or ideas. This is an important

factor as it means that CoPs can be formed from members of different social

categories or from different geographic regions. Wenger’s second component is

joint enterprise. Wenger states the importance the joint enterprise is constantly

67

renegotiated by the individual members. The joint enterprise goes beyond stated

goals (e.g., mission statement, objectives), but creates mutual accountability

among participants. The third component is a shared repertoire. “The

repertoire of a community of practice includes routines, words, tools, ways of

doing things, stories, gestures, symbols, genres, actions, or concepts that the

community has produced or adopted in the course of existence (Wenger, 1998a,

p. 83)

The above ideas are very much related to ideas from Brown & Duguid

(1991). In this seminal study, the authors state that the creation of knowledge

within communities of practice is characterized by three key elements. These

include:

1) Narratives: used for diagnosing problems and representing repositories of existing knowledge.

2) Collaboration: fuelled by participants engaged in and sharing

common practice

3) Social constructivism: participants develop a common understanding of their practice and of how to solve problems.

Brown & Duguid base their findings here primarily from ethnographic studies

undertaken by Orr (1987a; 1987b; 1990; 1990). These studies are important as,

according to Brown & Duguid, they help to illustrate how organizations depend

upon complex relationships between groups. Such relationships (as far as

organizations are concerned) do not formally exist, but may be most responsible

for community performance. Through these informal relationships, knowing is

validated and shared, and evolves through processes by individuals who engage

in the negotiation of meaning and through sharing insights and narratives (1991).

68

Another important idea, which relates to the notion of communities of

practice and social learning theory, is what Lave & Wenger (1991) call legitimate

peripheral participation (LPP). LPP is a type of situated learning, and is a

process that reiterates the focus that learning is fundamentally a social process

rather than solely psychological. Lave & Wenger support their theory through

observations of different apprenticeships (i.e., Yucatec midwives, Vai and Gola

tailors, US Navy quartermasters, meat-cutters, non-drinking alcoholics in

Alcoholics Anonymous). In these situations, people initially join communities and

learn from the periphery. As they become more competent they move closer to

the centre of each particular community. Thus, learning is not seen as the

acquisition of knowledge by individuals so much as a process of social

participation. The nature of the situation as the social context impacts

significantly on the process of learning and participation in the community.

Cultivating Communities of Practice

So now that we have identified the general characteristics of communities

of practice, now what? Much of the latest literature regarding CoPs is focused

less on describing these communities, and focused more on enabling these

informal communities to emerge, to flourish and to become productive. While, it

may be difficult to understand how organizations can promote something as

informal and voluntary as CoPs, Wenger, McDermott & Snyder (2000) use

examples as to how community structure can be encouraged. They write,

“Communities, unlike teams and other structures, need to invite the interaction to

69

make them alive. For example, a park is more appealing to use if its location

provides a short cut between destinations. It invites people to sit for lunch or chat

if it has benches set slightly off the main path, visible, but just out of earshot, next

to something interesting like a flower bed or a patch of sunlight” (p. 7). While of

course, this may seem simplistic, the example is meant to show that building

communities differs from contemporary organizational design which may

traditionally focus on creating structures, systems and roles toward achieving

specific organizational goals.

Wenger, McDermott & Snyder (2002) also set forth seven principles for

cultivating communities of practice, and in helping these communities gain what

they call “aliveness”. These principles, with paraphrased descriptions, follow:

1) Design for evolution: As CoPs are dynamic in nature, design should reflect adaptability (or the computer lingo term, scalability). The key to this point is to combine design elements that help to catalyze community development. “Physical structures – such as roads and parks – can precipitate the development of a town. Similarly, social and organizational structures, such as a community coordinator or problem-solving meetings, can precipitate the evolution of a community.”

2) Open a dialogue between inside and outside perspectives:

The authors state that good community design requires the perspective of an insider, one that is familiar with the types of activities within. However, the perspective of an outsider may help members see the possibilities within their own mechanisms, or in adopting other tools or procedures.

3) Invite different levels of participation: In any community,

there exist different levels of participation. While those on the peripheral may not participate in the same ways as those in the core, the peripheral members will still gain insights and knowledge through this type of participation. All members, regardless of participation levels, should be valued.

70

4) Develop both public and private community spaces: Members of communities interact with each other in both public and private functions. Thus, the public and private dimension of a community are interrelated. “The key to designing community spaces is to orchestrate activities in both public and private spaces that use the strength of the individual relationships to enrich events and use events to strengthen individual relationships.”

5) Focus on value: As communities are voluntary, value is key.

For members and prospective members, communities must offer value or there will may not be the incentive for participation. While value may not always be explicitly apparent, value should grow over time as the community evolves.

6) Combine familiarity and excitement: Familiarity, like the

comforts of a hometown, is important for a CoP. However, excitement is also as important, but in other ways. As communities mature, they settle into familiar ways of meeting and conduct. Yet, communities also need challenge and spontaneity to provide a break from everyday occurrences.

7) Creating a rhythm for the community: Like individuals’ lives

having a rhythm, “vibrant” communities also have a rhythm. “At the heart of a community is a web of enduring relationships among members, but the tempo of their interactions is greatly influenced by the rhythm of community events.” While all alive communities have a particular rhythm or tempo, it’s important to find the “right rhythm” at each stage of a community’s development.

(Wenger et al., 2002) Stages of Community Development Another important premise from Wenger, McDermott & Snyder (2002), is

what they refer to as the stages of community development. The authors have

identified three life phases, which include 5 stages. Collectively, these represent

the life cycle of a community. The stages and phases are described below.

71

1) Formation (potential and coalescing): Here, initial networks are discovered, common ground is formed and relationships are formed. The initial call (informally) is usually centred around the generation of value.

2) Integration (maturing and stewardship): At this stage, there

is a focus upon particular topics and the admission of new members. Tools and methods are developed that are unique to the community. New ideas are continually welcomed as the community evolves.

3) Transformation (transformation): At this stage, the

community may fade away or officially close. This may also mean that the community has become redundant, or that this stage brings about the beginning of a new community. Other possibilities include merging with other communities or becoming institutionalized as a formal unit.

Additionally, the authors have graphed the cycle of the community in

correspondence to the factors of time and energy/visibility. This is meant to be

representative of an average CoP life cycle. The diagram is included below.

Figure 5: Community Life Cycles Related to Time and Level of Energy and Visibility (Wenger et al., 2002)

72

Communities of Practice in the Online Environment Much recent literature has focused on the existence of online

communities. In this literature review, I have established some of the current

thought related to temporal communities, however, it is of my interest to also

examine the current thought regarding online community development and

communication. I found that much of the recent literature does not specifically

target CoPs, but also looks at virtual communities (VCs) or virtual learning

communities (VLCs). I believe a brief description of these types of communities

is also important. But first, a noteworthy thought regarding communities comes

from Brown (1999).

Community is quite possibly the most over-used word in the Net industry. True community – the ability to connect with people who have similar interests – may well be the key to the digital world, but the term has been diluted and debased to described even the most tenuous connections, the most minimal activity. (Online)

The important idea here is that it is important to understand the distinction

between online communication, and online community. I believe that many of

the characteristics from Wenger (1998) and Wenger et al. (2002) may help

researchers to make these important distinctions.

Two important ideas closely related to virtual communities include the

concepts of virtual learning communities (VLCs) and distributed communities of

practice (DCoPs). A definition of a VLC is “a group of people who gather in

cyberspace with the intention of pursuing learning goals” (Daniel, Schwier, &

McCalla, 2003). Alternately, a DCoP refers to a group of geographically

distributed individuals who are informally bound together by shared expertise and

73

shared interests or work (Daniel et al., 2003). The table below helps to identify

characteristics of such communities, as well as it helps to distinguish the concept

of a VLC from that of a DCoP.

Table 1: Key Features of Virtual Learning Communities and Distributed Communities of Practice (Daniel et al., 2003) Virtual Learning Communities (VLCs) Distributed Communities of Practice (DCoP)

- Less stable membership - Low degree of individual

awareness - More formalized and more

focused learning goals - More diverse language - Low shared understanding - Strong sense of identity - Strict distribution of

responsibilities - Easily disbanded - Low level of trust - Life span determined by extent

in which goals or requirements are satisfied

- Pre-planned enterprise and fixed goals

- Domain specific/interests

- Reasonably stable membership - High degree of individual awareness - Informal learning goals - Common language - High shared understanding - Loose sense of identity - No formal distribution of responsibilities - Less easily disbanded - Reasonable level of trust - Life span determined by the value the

community provides to its members - A joint enterprise is understood and

continually renegotiated by its members

- Shared practice/profession

The characteristics constructed above relate closely to Lave & Wenger’s

(1991) original framework regarding CoPs, but these specified features do much

more to describe shared and contrasting characteristics as they would apply in a

distributed environment. Key features of DCoPs include: shared interests,

common identity, shared information and knowledge, voluntary participation,

autonomy in setting goals, awareness of social protocols and goals, awareness

in membership and effective means of communication (Daniel et al., 2003).

Additionally, the important undertone that drives the entire community is

collaboration. Collaboration allows for the active exchange of ideas, and helps to

promote interest in being a part of the community.

74

Amy Jo Kim is another important figure it terms of designing communities

in online environments. Kim is the founder of Naima, a leading developer of

social architecture/online environments, and she has worked with various large

media companies (e.g., Sony, AOL, Yahoo) in designing online community

interfaces. Community Building on the Web (Kim, 2000) provides a simplistic,

yet comprehensive guide to the construction of online communities. Many of the

principles throughout this literature review, especially related to the

characteristics of CoPs, are reflected here, and Kim has adapted these principles

as they relate to her own practice and experience.

Kim (2000) organizes her thoughts around nine basic design principles

that have, thus far, characterized successful and sustainable online communities.

Together, these principles are developed as “social scaffolding” and are meant to

support and empower members. The principles are summarized below:

1) Define and articulate your PURPOSE: It’s important for members and prospective members to understand why the community is being built, and who it’s being built for. Be explicit through the design.

2) Build flexible, extensible gather PLACES: You should

develop a small-scale infrastructure of familiar gathering places. These will co-evolve through active membership.

3) Creating meaningful and evolving member PROFILES:

Profiles are important as they help to invoke communication between members, and help to give the community a sense of history and context.

4) Design for a range of ROLES: New members will have

different needs than senior members. Strategies around welcoming and empowering new members are important for those in leadership roles.

75

5) Develop a strong LEADERSHIP program: Community leaders are integral to the process as they greet and orient members to the community. It’s important that leaders are supported in these vital activities.

6) Encourage appropriate ETIQUETTE: While conflict can be

invigorating, it can also tear communities apart. Communities need to establish ground rules and conduct for communication processes.

7) Promote cyclic EVENTS: Events are important in instilling

rhythm to communities, as well as providing venues for socialization. Community leaders can establish events, or encourage members to set their own.

8) Integrate the RITUALS of community life: Rituals are

important in temporal communities, and may be as important in online communities. Rituals should be established around important occurrences (new members, exiting members, etc.)

9) Facilitate member-run SUBGROUPS: In large scale

communities, subgroups are very important as smaller groups can help to establish member loyalty and help to distinguish your community apart from others.

Kim (2000) does provide valuable strategies for designing online communities.

Additionally, these guidelines integrate well with the previous literature in

understanding distinguishing characteristics of communities of practice and

virtual learning communities.

Conclusion To Communities of Practice The concept of communities of practice is an important one when

attempting to understand the complex relationships found between individuals in

both temporal and online environments. While this is a only a small sample of

the field, it is representative of some of the major thinkers and theories which

underlie this emerging concept. I believe that any study focusing on the

76

individuals, groups and relationships forming open source communities must

consider CoPs literature.

Social Capital Theory

We’ve all heard the common saying, “it’s not what you know, it’s who you

know.” In a simplistic sense, this aphorism does well to capture much of the

essence of what theorists explain as social capital theory (SCT). Social capital

(SC), in the loosest sense, is a term used to describe the social networks that

exist between individuals and groups, and is often used to measure the character

and relative strengths of the ties that exist. While the concept of social capital

has existed since the early 1900’s (attributed to LJ Hannifin), it seems to have

gained greater relevance to theorists in the last several decades. (Sander &

Lowney, 2003; Woolcock & Narayan, 2000).

There are various descriptions of social capital used throughout the

literature. Here are a few definitions of SC to consider.

… those tangible substances [that] count for most in the daily lives of people.42 (Hannifin, 1916, p. 130) The aggregate of the actual or potential resources which are linked to possession of a durable network of more or less institutionalized relationships of mutual acquaintance or recognition. (Bourdieu, 1986, p. 243) … the sum of the actual and potential resources embedded within, available through, and derived from the network of relationships possessed by an individual or social unit. (Nahapiet & Ghoshal, 1998, p. 243)

42 At the time, Hannifin was particularly focused upon good will, fellowship, sympathy and social intercourse among those that make up a ‘social unit’. (http://www.infed.org/biblio/social_capital.htm)

77

… refers to the institutions, relationships, and norms that shape the quality and quantity of society’s social interactions…. Social capital is not just the sum of the institutions which underpin a society – it is the glue that holds them together. (The World Bank, 1999) Whereas physical capital refers to physical objects and human capital refers to the properties of individuals, social capital refers to connections among individuals – social networks and the norms of reciprocity and trustworthiness that arise from them. (Putnam, 2000, p. 19) … refers to the norms and networks that enable people to act collectively. (Woolcock & Narayan, 2000, p. 226) … consists of the stock of active connections among people: the trust, mutual understanding, and shared values and behaviors that bind the members of human networks and communities and make cooperative action possible. (Cohen & Prusak, 2001, p. 4)

While there are many other definition of social capital,43 the previous

definitions are fairly representative of those that exist in the literature. There are

certainly crossovers and commonalities, but a few slight discrepancies exist as

well. Nahapiet & Ghaoshal (1998) provide a useful framework in describing social

capital. The authors express SC in terms of three primary dimensions. First, they

propose that social capital describes the intricate connections between

individuals and social groups. Individuals may perceive themselves as part of a

network or social structure in which they live, work and learn. In order to perform

daily functions, advantages are often gained through the understanding and

leveraging of relationships. This is what is known as the structural dimension.

Second, trust and reciprocity are important binding ties and may allow individuals

to work more openly with each other. Thus, the corollary, individuals who work in

trusting relationships are likely to work more effectively. Nahapiet & Ghaoshal

43 For an extensive list of definitions, see (Adler & Kwon, 2000)

78

(1998) identify this as the relational dimension. Third, members of the network

may have common interests or shared understandings related to their

organizational structure (formal or informal). And central to this cognitive

dimension, members of the community partake in knowledge building and

sharing. With this framework, each dimension is interdependent of the others,

and the strength of one dimension may positively affect another.

Social capital has been studied from many perspectives and disciplines.

Social capital research has been particularly intense in the areas of management

and organizational behaviour, economic development and in the study of civic

engagement. Studies from these areas provide an excellent basis for social

capital theory, but I am particularly interested in social capital as it exists in

distributed communities of practice such as an open source community. For this,

a definition of social capital as it relates to virtual learning communities is helpful.

Social capital in VLCs are defined “as a common social resource that facilitates

information exchange, knowledge sharing, and knowledge construction through

continuous interaction, built on trust and maintained through shared

understanding” (Daniel et al., 2003, Online). This definition does well to convey

the three relational dimensions outlined by Nahapiet & Ghaoshal (1998), and

focuses on individual/organizational learning through knowledge activities.

Although offered as a description of social capital in VLCs, in my opinion, this

definition works well to describe social capital as it exists in open source

communities.

79

Types of Social Capital

Like physical capital, social capital is not a monolithic entity and

encompasses a broad range of items that may be described as types of SC.

Sander & Lowney (2003, p. 4) have developed four categories in which social

capital can be divided. These types of social capital include:

1) Public-regarding vs. private regarding: Community members may form around various issues – some public (e.g., Parent Teacher Association), and some quite personal (e.g., postnatal support groups). Sander & Lowney (2003) hypothesize that the latter may attract more community members (but smaller groups), yet the former may be more action-oriented toward solving particular community issues.

2) Formal vs. Informal: Communities are formed for various reasons, by

various forces. However, some of the most intricate communities may be those that began as formal communities and slowly evolved into an informal entity. While formal ties may better ensure longevity of a community, informal ties may help to deepen relational bonds.

3) Bridging vs. Bonding: Bridging types of social capital may focus on

tying individuals across barriers of race, class or ethnicity. Bonding types may focus on strengthening existing social relationships. While these acts may occur simultaneously, Sander & Lowney (2003) believe that bonding communities may be easier to build, but bridging communities are important in build a sense of unity across diversity.

4) Strong Ties vs. Weak Ties: Strong ties enable the discussion and

production of different types of knowledge than which is developed by weaker ties. For instance, Sander & Lowney (2003) hypothesize that members of a community may be enable to discuss serious and sometimes personal issues (e.g., serious health problems, marital problems) with those whom we have strong ties with, while the discussion of less serious topics (e.g., a one day park cleanup) are a limitation of weaker ties. The strength of community ties will have an effect on what a community can do collectively (Wuthnow, 1998).

While it is not made explicit how each type of social capital will affect a

community in every case, it is important to pay attention to what types of capital

80

are paid attention to. This is especially important if one is using the idea of social

capital to measure community ties, or attempting to build communities.

Measuring Social Capital

Much of the drive to understand the concept of social capital is related to

the attempt to improving it, or in other words, to develop greater social capital

within communities. With this assumption, theorists move from understanding

social capital to an attempt to measure it using both quantitative and qualitative

techniques. Yet, there is a risk in classifying aspects of human relationships as

“capital”. Cohen & Prusak (2001, p. 9) warn, “there is a deep danger of skewing

our consideration of social phenomenon and goods toward the economic. The

notion of capital bring with it a whole set of discourses and inevitably links it, in

the current context, to capitalism.” In understanding social capital, its value and

how it may or may not be measured, this is an important principle to consider.

There are various voices in the literature offering techniques and tools for

measuring elements of social capital. However, one does not have to wade very

far to see examples that disappoint. The publication Measuring Social Capital: A

Guide for First Nations Communities (Mignone, 2003), features a questionnaire

tool which attempts to gain perspective into the connected or disconnected lives

of First Nations community life. Sample questions, as represented in Figure 6

below, are used to measure degrees of ‘bonding’ activities in the community.

While the design of the tool attempts to nurture understanding of the types of

social relationships that exist in First Nations’ communities, the largely

81

quantitative approach may miss the target with its assumptive questions and

focus on activity counting, rather than activity description.

Figure 6: Bonding Targeted Questions for Measuring Social Capital

(Mignone, 2003, p. 8)

There have been many more attempts at designing tools for measuring

social capital. Notable quantitative studies include work by Knack & Kiefer (1997)

who used indicators of trust and civic norms in sampling 29 market economies.

Additionally, Temple & Johnson (1998) examined the influences and indicators

of social capital such as ethnic diversity and social mobility in relation to the

concentration of telephone services in several sub-Saharan African countries.

Select qualitative studies include Portes & Sensenbrenner (1993) who examined

the effects on immigrant communities when certain members of the community

experienced economic success and wished to leave their communities. Also of

note, Fernandez-Kelley (1995) studied young girls in urban ghetto communities

and discovered that normative pressures toward teen child-rearing and the

82

rejection of formal employment were powerful forces in the community. While, I

will not go into the pros and cons of performing qualitative vs. quantitative studies

in this paper, I do see the richness of data gathered through qualitative

techniques and the power of the narrative that develops. In measuring social

capital, researchers must be sensitive that they are attempting to understand

social relationships and measures of trust within communities. While quantitative

approaches do have some merit, qualitative methodologies are likely to give us a

much richer perspective.

How Does This All Fit Together?

For a moment I am going to leave the world of open source and attempt to

use the knowledge I have gained so far (i.e., change theory, communities of

practice, social capital theory) in another context. My thoughts here are inspired

by real live events, and from the experiences of a supportive colleague.

A Baby Story: Expecting Parents

Claudia and Alec are expecting a baby. As first time parents to be, they

are quite nervous about preparing for the arrival of their child, and are looking for

as much information and support as they can find. Searching the Internet, they

have found many excellent resources such as BabyCentre.com which provides

up-to-date relevant information on the expected baby’s development. Yet still,

resources like this seem to lack the human element that is preferable to a life

changing event such as this. Alec heard about perinatal classes held at the

83

YMCA, and was told excellent things about the instructor, Sally Elliot. With little

hesitation, Alec and Claudia signed up for the classes.

When arriving at the “Y”, although Alec realized that these were perinatal

classes, he was shocked to see this many pregnant women, and by association,

this many expecting fathers in the same room. Alec and Claudia could see that

they were not alone in this experience, and that there were many others in the

room that (obviously) shared the same reasons for attending classes. This was a

group that came together through similar common goals: to gain knowledge

about the processes involved in childbirth, to learn about health and safety issues

regarding the baby and mother and to gain a better understanding of how to

prepare their child for life beyond the womb.

Sally was an amazing instructor, and the knowledge gained from the

course was incredibly valuable. Sally was also very approachable and took the

time to be supportive to each couple individually. However, as the group was

often nervous, and unfamiliar with each other, Sally had a commanding presence

in the classroom and took a very directed approach to teaching. Much was

learned from the class, yet there was little done to build relationships between

the members. The common tie that built this ‘community’ was based on the first-

birth experience, and likely the fear and uncertainty that goes with this first

experience. However, beyond this tie, there was little opportunity to explore other

shared commonalities of the group members.

84

A Baby Story: A New Mother (based on a faculty colleague)

Jane is a new mother. Jane and her husband Tom had attended Sally’s

classes and had enjoyed and learned much from these experiences. However,

Jane now hoped for new experiences and connections that would help her

understand what it meant to become a good mother. In Sally’s class, she had

heard about a group of new mother’s called “Wise Moms” which is a post-partum

support group. Jane was a bit hesitant and nervous, but was soon a member of

the “Wise Moms” group.

As expected, the structure and protocols of this group were much different

than what was experienced in Sally’s formal classes. There was no particular

curriculum or agenda, but the new mothers simply met to share experiences,

voice concerns, ask questions and share answers concerning what it means to

be a new mother. Sally also took a different role in the group. Instead of

determining the direction of the class, the participants now had much more

control of the discussed content. Whereas Sally “wore the hat” of a registered

nurse and instructor in the classes, Sally developed her role in the ‘Wise Moms’

group from the experience of a mother. With this more informal structure, roles

were constantly being negotiated. At times, certain voices were made more

relevant than others. For instance, Jane remembers questioning the credibility of

the mother that let her 3-month old baby drink Slurpies, while Jane gave full

attention to the member who was a second-time mother and practicing registered

nurse.

85

As her babies grew into young children, Jane quit the group as she felt

she no longer had a need to participate in it. She was no longer ‘getting anything

from it’. However, the experiences in “Wise Moms” led to the development of

strong bonding relationships, and excellent lasting friendships. While at one time,

being a new mother was the only commonality shared with the individual

mothers, the group facilitated an exchange that allowed other shared beliefs and

values to be known to each other. Today, “Wise Moms” continues to seed

relationships that last for years after the group participation.

So What Does This All Mean?

The specific characteristics of open source communities vary. While many

open source communities exist as informal, distributed entities, there have been

attempts at developing more formal, proximal open source communities (e.g.,

local Linux User Groups or LUGs). As well, within many existing open source

communities, there are aspects of more formalized and direct approaches to

educating members (and prospective members) on the virtues and technicalities

of open source related software (e.g., installfests, school staff workshops). In

many ways, open source communities share some of the characteristics of the

‘baby’ communities presented above, and can be just as diverse in structure,

purpose and in the facilitation of relationships.

In using the lenses of change theory, communities of practice theory and

social capital theory, I hope to gain insight into the technology adoption practices

of particular members of OS communities. As open source communities are

86

complex, one lens is often not adequate to explain the activities and relationships

within these communities. The weakness of one model can benefit from the

strength of another. For instance, change theory can be very powerful when

describing directed approaches to change. For instance, in the case of “Baby

Story: New Parents”, change theory such as diffusion theory may be appropriate

in understanding basic components of the class experiences. Yet, diffusion

theory would not likely be appropriate in understanding the experiences and

relationships developed in the second vignette, “A Baby Story: A New Mother”. In

this case, having a solid understanding of communities of practice and social

capital literature may give a much richer understanding of the context. While

perhaps using too many theories may be unwise, it is my belief that in using the

grounded theory process, the more tools I have access to will only create a richer

analysis of the data.

Visualizing My Framework

Before presenting my framework, I would like to re-present some of the

key ideas that have helped in my understanding. Key words have been bolded in

an effort to assist in the development of my visual model.

Communities of Practice Defined

… a group of professionals informally bound to one another through exposure to a common class of problems, common pursuits of solutions, and thereby themselves embodying store of knowledge. (Hildreth & Kimble, 2000, p. 3) … groups of people who share a concern, set of problems, or a passion about a topic, and who deepen their knowledge and expertise in the area by interacting on an ongoing basis. (McDermott & Snyder, 2002, p. 7)

87

Social Capital Defined

… a common social resource that facilitates information exchange, knowledge sharing, and knowledge construction through continuous interaction, built on trust and maintained through shared understanding (Daniel, et. al, 2003, Online) … consists of the stock of active connections among people: the trust, mutual understanding, and shared values and behaviours that bind the members of human networks and communities and make cooperative action possible. (Cohen & Prusak, 2001, p. 4)

In understanding the adoption of technological innovation in open source

communities, it’s important to recognize that planned or institutional adoption

techniques are only one of many forces influencing potential adopters. From my

current knowledge, educators who are pioneers in the open source movement

have not been solely inspired to be so due to institutional initiatives. Rather, their

involvement in various overlapping open source communities have inspired such

activities and given them opportunities to learn difficult technical skills, as well as

to forge strong (often philosophical) beliefs regarding the meaning of open

source in the educational context.

Below (Figure 7) you will see a (very) rudimentary illustration44 of various

thoughts so far. As you can see, I feel that communities of practice and attributes

of social capital are closely related. In fact, I see these attributes of social capital

as an extension and direct result of effective communities of practice.

Additionally, in being realistic towards this research, I must admit that some

elements of rich, active communities (to which I have labelled as “nebulous”)

cannot be easily described through change related theories such as diffusion of

innovation theory. However, such theories do serve a purpose and may be 44 Forgive my lacking of artistic ability.

88

beneficial in helping us better understand particular aspects of open source

communities.

Figure 7 : Communities of Practice in Relation to Social Capital Attributes of Social Capital Knowledge Sharing Activity / Interaction Relationships Connections Shared Values Common Context/Goals Trust

Change Theory helpful in describing predictable/planned change.

Social Capital Theory and Communities of Practice theory helpful in understanding nebulous activity.

As mentioned previously, the above thoughts are meant only as a

guiding framework for the time being. In keeping with the spirit of grounded

theory, I expect to be immersed in the related and likely tangential literature

throughout the course of my study. I have kept a well rounded approach toward

my literature research, and thus, I am hoping to keep this framework as flexible

as possible. The areas of change theory, communities of practice and social

capital are exciting areas of research, and I believe that my thorough

understanding of these areas can only be beneficial to understanding open

source communities.

89

Bibliography

Abrahamson, E., & Rosenkopf, L. (1997). Social network effects on the extent of innovation diffusion: A computer simulation. Organization science, 8, 289-309.

Adler, P., & Kwon, S. (2000). Social capital: The good, the bad, and the ugly. In

E. P. Lesser, L. (Ed.), Knowledge and social capital. Woborn, MA: Butterworth Heinemann.

Arnison, M. (2001). Open publishing is the same as free software. Available from:

http://www.cat.org [Accessed 08 June 2003] Atlas.ti. (1997). Visual qualitative data analysis management model building in

education, research & business. Retrieved May 1, 2004 Bandura, A. (1977). Social learning theory. Englewood Cliffs, N.J.: Prentice-Hall. Bank, T. W. (1999). What is social capital? Retrieved February 12, 2004, from

http://www.worldbank.org/poverty/scapital/whatsc.htm Barbrook, R. (1998). The hi-tech gift economy. First Monday. [online] Available

from: http://www.firstmonday.dk/issues/issue3_12/barbrook/ Bates, T. (2000). Managing technological change : strategies for college and

university leaders (1st ed.). San Francisco: Jossey-Bass. Bennahum, D. (1996). Interview with Richard Stallman. Meme 2.04. [online]

Available from: http://hammer.prohosting.com/~runlinux/stallman.shtml [Accessed 07 June 2003]

Bourdieu, P. (1986). The forms of capital. In J. G. Richardson (Ed.), Handbook of theory and research for the sociology of education (pp. 241-258). New York: Greenwood Press.

Bowling, A. (1997). Research Methods in Health. Buckingham: Open University

Press. Brown, J. S. (1999). There goes the neighborhood. Retrieved November 5, 2003,

from http://archive.salon.com/21st/feature/1999/01/cov_19feature.html Brown, J. S., & Duguid, P. (1991). Organizational learning and communities-of-

practice: Toward a unified view of working, learning, and innovation, from http://www2.parc.com/ops/members/brown/papers/orglearning.html

Cheal, D. (1988). The gift economy. New York: Routledge.

90

Cohen, D., & Prusak, L. (2001). In good company. How social capital makes organizations work. Boston, Ma: Harvard Business School Press.

Coleman, G. E. (1999). The poltics of surival and prestige: Hacker indentity and

the global production of an operating system. Retrieved July 14, 2001, from http://www.healthhacker.com/biella/masterslongversion.html

Coyne, I. (1997). Sampling in qualitative research: Purposeful and theoretical

sampling: merging or clear boundaries. Journal of Advanced Nursing, 26, 623-630.

Creswell, J. W. (1998). Qualitative inquiry and research design : choosing among

five traditions. Thousand Oaks, Calif.: Sage Publications. Daniel, B., Schwier, R. A., & McCalla, G. (2003). Social capital in virtual learning

communities and distributed communities of practice. Canadian Journal of Learning and Technology, 29(3).

Daniel, B., McCalla, G., & Schwier, R. (2002). A process model for building social

capital in virtual learning communities. Proceedings of the International Conference on Computers in Education (ICCE) (pp., 574-577). Auckland, New Zealand.

Dent, H. S. (1994). The great boom ahead: Your guide to personal & business

profit in the great age of prosperity. New York: Hyperion Press. Denzin, N. K., & Lincoln, Y. S. (1998). Strategies of qualitative inquiry. Thousand

Oaks, Calif.: Sage Publications. Denzin, N. K., & Lincoln, Y. S. (Eds.). (1994). Handbook of qualitative research.

Thousand Oaks: Sage Publications. Dey, I. (1999). Grounding grounded theory : guidelines for qualitative inquiry. San

Diego: Academic Press. Dibona, C., Ockman, S. & Stone, M. (Eds.) (1999). Open sources: Voices from

the open source revolution. Sebastopol, CA: O'Reilly & Associates. Dick, B. (2002, December 12, 2002). Grounded theory: a thumbnail sketch.

Retrieved February, 2004, from http://www.scu.edu.au/schools/gcm/ar/arp/grounded.html Eisenhardt, K. M. (1989). Building theories from case study research. Academy

of management review, 14, 532-550.

91

Ellsworth, J. (2000). Surviving change: A survey of educational change models. ERIC.

Ely, D. (1990a). Conditions the facilitate the implementation of educational

technology innovations. Journal of Research on Computing in Education, 23(2), 298-305.

Ely, D. (1990b). The diffusion and implementation of educational technology in

developing nations: Cross-cultural comparisons of Indonesia, Chile, and Peru. Instructional Developments, 1(1), 9-12.

Ely, D. (Ed.). (1976). Creating the conditions for change. Champaign, IL:

University of Illinois Graduate School of Library Science. Fernandez-Kelley, M. P. (1995). Social and cultural capital in the urban ghetto:

Implications for the economic sociology of immigration. In A. Portes (Ed.), The economic sociology of immigration. New York: Russel Sage Foundation.

Fink, M. (2003). The business and economics of linux and open source. New

York: Prentice Hall. Free Software Foundation (2003). The Free Software Definition. [online]

Available from: http://www.gnu.org/philosophy/free-sw.html [Accessed 08 June 2003]

Fullan, M. (2001). The new meaning of educational change (3rd ed.). New York:

Teachers College Press. Fullan, M., & Miles, M. B. (1992). Getting reform right: What works and what

doesn't. Phi Delta Kappan, 73(10), 745-752. Fullan, M., & Stiegelbauer, S. M. (1991). The new meaning of educational

change (2nd ed.). New York, NY: Ontario Institute for Studies in Education: Teachers College Press Teachers College Columbia University.

Gladwell, M. (1996, June 3). The Tipping Point. Retrieved July 7, 2003, from

http://www.gladwell.com/1996/1996_06_03_a_tipping.htm Gladwell, M. (2000). The tipping point : how little things can make a big

difference (1st ed.). Boston: Little Brown. Glaser, B. G. (1978). Theoretical Sensitivity. Mill Valley, CA: Sociology Press.

92

Glaser, B. G. (1992). Basics of grounded theory analysis: Emergence vs. forcing. Mill Valley, CA.

Glaser, B. G., & Strauss, A. L. (1967). The discovery of grounded theory :

strategies for qualitative research. Chicago: Aldine Pub. Co. Glesne, C. (1999). Becoming qualitative researchers : an introduction (2nd ed.).

New York: Longman. Gordon, G. (2002). The roles of leadership and ownership in building an effective

quality culture. Quality in Higher Education, 8(1), 97-106. Hall, G. E., & Hord, S. M. (1987). Change in schools : facilitating the process.

Albany, N.Y.: State University of New York Press. Hannifin, L. J. (1916). The rural school community centre. Annals of the

American Academy of Political and Social Science, 67, 130-138. Hargreaves, D. H. (2003). Working laterally: How innovation networks make an

education epidemic., from http://www.demos.co.uk/workinglaterally Havelock, R. G., & Zlotolow, S. (1995). The change agent's guide (2nd ed.).

Englewood Cliffs, N.J.: Educational Technology Publications. Hildreth, K. (2000). Communities of Practice in the Distributed International

Environment. (4), 17. Hildreth, P., Kimble, C., Wright, P. (2000). Communities of practice in a

distributed international environment. Journal of Knowledge Management, 4(1), pp. 27-38.

Hyde, G. (2000). Independent media centers: Cyber-subversion and the

alternative press. First Monday. [online] Available from: http://www.firstmonday.dk/issues/issues/issue7_4/hyde

Kerlin, B. (2002). Coding Strategies. Retrieved February 14, 2004, from

http://kerlins.net/bobbi/research/nudist/coding/strategies.html Kim, A. J. (2000). Community building on the web: Secret strategies for

successful online communities. Berkeley, CA: Peachpit Press. Kim, E. (2003). An introduction to open source communities. [online]. Available

from: http://www.blueoxen.org/research/00007 [Accessed 07 July 2003]

93

King, J.J. (1999). Free Software is a Political Action: In Conversation With Richard Stallman. Teleopolis. [online] Available from: http://www.heise.de/tp/english/special/wos/6469/1.html [Accessed 07 June 2003]

Knack, S., & Kiefer, P. (1997). Does social capital have an economic payoff? A

cross-country investigation. Quarterly Journal of Economics, 112, 1251-1288.

Knight, P., & Trowler, P. R. (2001). Departmental leadership in higher education.

Philadelphia: Open University Press. Kvale, S. (1996). Interviews : an introduction to qualitative research interviewing.

Thousand Oaks, Ca.: Sage Publications. Lave. (1988). Cognition in practice. Cambridge, UK: Cambridge University Press. Lave, J., & Wenger, E. (1991). Situated learning: legitimate peripheral

participation: Cambridge University Press. Lee, R. M., & Fielding, N. G. (1991). Computing for qualitative research: Options,

problems and potential. In N. G. Fielding (Ed.), Using computers in qualitative research (pp. 1-13). London: Sage.

Lerner, J. & Tirole, J. (2000). The simple economics of open source. Harvard

Business School working paper #00-059. [online] Available from: http://www.people.hbs.edu/jlerner/simple.pdf [Accessed 04 August 2003]

Lessig, L. (2004). Free culture: How big media uses technology and the law to

lock down culture and control creativity. New York: Penguin Press. Mauss, M. (1990). The gift: The form and reason for exchange in archaic

societies. New York: Routledge. May, C. (2000). Global political economy of intellectual property rights: The new

enclosure? London: Routledge. McGriff, S. (2001, October 27, 2000). ISD Knowledge Base / Educational

Change Models. Retrieved January 21, 2004, from http://www.personal.psu.edu/faculty/s/j/sjm256/portfolio/kbase/Systems&Change/EducationalChangeModels.html

McGowan, D. (2001). The legal implications of open source software. Illinois Law

Review. 2001(1). [online]. Available from: http://www.law.umn.edu/FacultyProfiles/articles/OpenSourceFinal.pdf [Accessed 08 July 2003]

94

Midgley, D. F., & Dowling, G. R. (1978). Innovativeness: The concept and its measurement. Journal of Consumer Research, 4, 229-242.

Mignone, J. (2003). Measuring social capital: A guide for first nations

communities. Retrieved February 12, 2004, from http://secure.cihi.ca/cihiweb/en/downloads/MeasuringSocialCapital2003_e.pdf

Moglen, E. (1999). Anarchism triumphant: Free software and the death of

copyright. First Monday. [online]. Available from: http://emoglen.law.columbia.edu/my_pubs/anarchism.html [Accessed 09 June 2003]

Moore, J.T.S. (Producer/Writer/Director). (2002). Revolution OS [Motion Picture].

United States: Wonderview Productions. Nahapiet, J., & Ghoshal, S. (1998). Social capital, intellectual capital and the

organizational advantage. Academy of Management Review, 23(2), 242-266.

Newmarch, J. (2001). Lessons from open source: Intellectual property and

courseware. First Monday. [online] Available from: http://www.firstmonday.dk/issues/issue6_6/newmarch/ [Accessed 10 July 2003]

Orr, J. (1987a). Narratives at work: Story telling as cooperative diagnostic

activity. Field Service Manager, 47-60. Orr, J. (1987b). Talking about machines: Social aspects of expertise. Palo Alto,

CA: Xerox Palo Alto Research Centre. Orr, J. (1990). Collective remembering: Memory in society. In D. Edwards (Ed.),

Sharing knowledge, celebrating identity: War stories and community memory in a service culture. Beverley Hills, CA: Sage Publications.

Orr, J. (1990). Talking about machines: An ethnography of a modern job.

Unpublished Ph.D., Cornell University. Patton, M. Q. (1990). Qualitative evaluation and research methods (2nd ed.).

Newbury Park, Calif.: Sage Publications. Patton, M. Q. (2002). Qualitative research and evaluation methods (3rd ed.).

Thousand Oaks, Calif.: Sage Publications. Petroski, H. (1992). The evolution of useful things. New York: Random House.

95

Portes, A., & Sensenbrenner, J. (1993). Embeddedness and immigration: Notes on social determinants of economic action. American Journal of Sociology, 98(6), 1320-1350.

Putnam, R. D. (2000). Bowling alone : the collapse and revival of American

community. New York: Simon & Schuster. Ramsden, P. (1998). Learning to lead in higher education. New York: Routledge. Raymond, E. (1999). The cathedral and the bazaar: Musings of linux and open

source by an accidental revolutionary. Sebastopol, CA: O’Reilly & Associates Inc. Raymond, E. (1998). Homesteading the noosphere. First Monday. [online] Available http://www.firstmonday.dk/issues/issue3_10/raymond

Reigeluth, C. M., & Garfinkle, R. J. (1994). Systemic change in education.

Englewood Cliffs, N.J.: Educational Technology Publications. Rogers, E. M. (1995). Diffusion of innovations (4th ed.). New York: Free Press. Rogers, E. M., & Kincaid, D. L. (1981). Communication networks : toward a new

paradigm for research. New York: Free Press. Rubin, H. J., & Rubin, I. (1995). Qualitative interviewing : the art of hearing data.

Thousand Oaks: Sage Publications. Ryan, B., & Gross, N. (1943). The diffusion of hybrid seed corn in two Iowa

communities. Rural Sociology, 8(1), 15-24. Sander, T. H., & Lowney, K. (2003). Social Capital Building Toolkit (Version 1.0).

Retrieved December 23, 2003, from www.nelib.org/files/counterpartsSkillbuildingToolkit.pdf

Seidman, I. (1998). Interviewing as qualitative research : a guide for researchers

in education and the social sciences (2nd ed.). New York: Teachers College Press.

Sodenberg, J. (2002). Copyleft vs. copyright: A Marxist critique. First Monday.

[online] Available from: http://emoglen.law.columbia.edu/my_pubs/anarchism.html [Accessed 07 June 2003]

96

Stallman, R. (2000). The GNU operating system and the free software movement. [online] In C. DiBona, S. Ockman, and M. Stone (Eds.) Open Sources: Voices from the Open Source Revolution. Sebastopol, CA: O'Reilly & Associates. Available from: http://www.oreilly.com/catalog/opensources/book/stallman.html [Accessed 07 June 2003]

Strauss, A. L. (1987). Qualitative analysis for social scientists. Cambridge

[Cambridgeshire] ; New York: Cambridge University Press. Strauss, A. L., & Corbin, J. M. (1990). Basics of qualitative research : grounded

theory procedures and techniques. Newbury Park, Calif.: Sage Publications.

Strauss, A. L., & Corbin, J. M. (1997). Grounded theory in practice. Thousand

Oaks: Sage Publications. Tarde, G. d., & Parsons, E. W. C. (1903). The laws of imitation. New York,: H.

Holt and company. Temple, J., & Johnson, P. (1998). Social Capability and Economic Growth.

Quarterly Journal of Economics, 113(3), 965-990. Tesch, R. (1990). Qualitative research: Analysis types and software tools. New

York: Falmer Press. Tesch, R. (1991). Software for qualitative researchers: Analysis needs and

program capabilities. In R. M. Lee (Ed.), Using computers in qualitative research (pp. 16-37). London: Sage.

Torvalds, L. (1991, August 25). What would you like to see most in minix?

Message posted to comp.os.minix electronic mailing list, archived at http://groups.google.ca [Accessed 03 June 2003]

Torvalds, L. & Diamond, D. (2001). Just for fun: Story of an accidental

revolutionary. New York: Harper Collins. Valente, T. W. (1995). Network models of the diffusion of innovations. Cresskill,

NJ: Hampton Press). Vygotski, L. S., & Cole, M. (1978). Mind in society : the development of higher

psychological processes. Cambridge: Harvard University Press. Weber, S. (2004). The success of open source. Cambridge: Harvard University

Press.

97

Wenger, E. (1998a). Communities of practice : learning, meaning, and identity. Cambridge, U.K. ; New York, N.Y.: Cambridge University Press.

Wenger, E. (1998b). Communities of Practice. Learning as a Social System,

2002, from http://www.co-i-l.com/coil/knowledge-garden/cop/lss.shtml Wenger, E., McDermott, R. A., & Snyder, W. (2002). Cultivating communities of

practice : a guide to managing knowledge. Boston: Harvard Business School Press.

Wenger, E. (1998). Communities of practice: Learning, meaning and identity.

Cambridge: Cambridge University Press. Wenger, E., Snyder, M. (2000). Learning within communities. [online] Available

from: http://www.linezine.com/1/features/ewwslc.htm [Access 11 June 2003]

Wheeler, D. (2003). Why open source software / free software (OSS/FS)? Look

at the numbers! [online] Available from: http://www.dwheeler.com/oss_fs_why.html [Accessed 10 June 2003]

Wiley, D. (1998). Open Content [online] Available from:

http://www.opencontent.org [Access 22 May 2003] Wengraf, T. (2001). Qualitative research interviewing : biographic narrative and

semi-structured methods. London ; Thousand Oaks, Calif.: Sage. Woolcock, M., & Narayan, D. (2000). Social capital: Implications for development

theory, research, and policy. The World Bank Research Observer, 15(2), 225-249.

Wuthnow, R. (1998). Loose connections : joining together in America's

fragmented communities. Cambridge, Mass.: Harvard University Press. Zaltman, G., & Duncan, R. (1977). Strategies for planned change. New York:

Wiley.

98

Appendix A

S-Curve of the Automobile

An example of an S-Curve is shown below:

At the beginning of the 20th century, only the very rich owned an automobile.

Following this S-Curve, between 1900 and 1914, the automobile went through

the innovation phase. At the end of this phase, Henry Ford introduced the

assembly line which helped the automobile become affordable for the middle

class. From 1914-1928, the automobile went through it’s growth phase as 90%

of urban families now owned one, up drastically from 10%, only 14 years earlier.

After 1928, the automobile market grew slowly as it reached the maturity stage.

(Paraphrased from Dent, 1994, pp. 106-8)

99

Appendix B

The CBAM Model: Stages of Concern

(Hall and Hord, 1987)

100

Appendix C

Initial ideas of social learning theory are attributed to the work of Bandura

in the late 1970’s. Bandura (1977) emphasized the importance of observing and

modelling the behaviours, attitudes and emotional reactions of others. Bandura

(1977) believed that “most human behavior is learned observationally through

modeling: from observing others, one forms an idea of how new behaviours are

performed, and on later occasions this coded information serves as a guide for

action” (p. 22). Bandura constructed social learning theory as both a

behaviourist and cognitive model as he used it to explain human action in terms

of a continuous reciprocal interaction between behavioural, cognitive and

environmental influences. In many ways, Bandura’s work complimented ideas

from Vygotsky’s (1978) theory that social interaction plays a fundamental role in

the development of cognition and Lave’s (1988) theory of situated learning.